Abstract
Background
Medial clavicle fractures are uncommon, occurring in older and multiply injured patients. The management of these fractures and the factors that predispose toward poor outcomes are controversial. Furthermore, the functional outcomes of treatment are not well characterized or correlated with fracture patterns.
Questions/purposes
(1) To determine minimum 1-year functional outcomes using QuickDASH scores and pain scores after medial clavicle fractures and (2) to identify factors associated with these outcome variables.
Methods
In an institutional review board–approved, retrospective study, we identified adult patients with medial clavicle fractures at two tertiary care referral centers in a single metropolitan area in the United States from January 2010 to March 2019. Our initial query identified 1950 patients with clavicle fractures, from which 74 adult patients with medial clavicle fractures and at least 1 year of follow-up were identified. We attempted to contact these eligible patients by telephone for functional outcomes and pain scores. Twenty-six patients were deceased according to the most recent Social Security Death Index data and public obituaries, three declined participation, and 14 could not be reached, leaving 42% of the total (31 of 74) and 65% (31 of 48) of living patients included in the analysis. Demographic characteristics, fracture characteristics, and clinical and radiographic union as assessed by plain radiography and CT were collected through record review. Twenty-nine patients were treated nonoperatively and two patients underwent open reduction internal fixation. Sixty-eight percent (21 of 31) of the included patients also had radiographic follow-up at least 6 weeks postoperatively; two patients had persistent nonunion at a mean of 5 ± 3 years after injury. Our primary response variable was the QuickDASH score at a minimum of 1 year (median [range] 5 years [2 to 10]). Our secondary response variable was the pain score on a 10-point Likert scale. A bivariate analysis was performed to identify factors associated with these response variables. The following explanatory variables were studied: age, gender, race, dominant hand injury, employment status, manual labor occupation, primary health insurance, social deprivation, BMI, diabetes mellitus, smoking status, American Society of Anesthesiologists physical status classification, Charlson Comorbidity Index, nonisolated injury, high-energy mechanism of injury, nondisplaced fracture, fracture comminution, superior-inferior fracture displacement, medial-lateral fracture shortening, and surgical treatment of the medial clavicle fracture.
Results
The mean QuickDASH score was 12 ± 15, and the mean pain score was 1 ± 1 at a mean of 5 ± 3 years after injury. The mortality rate of the cohort was 15% (11 of 74) at 1 year, 22% (16 of 74) at 3 years, and 34% (25 of 74) at 5 years after injury. With the numbers available, no factors were associated with the QuickDASH score or pain score, but it is likely we were underpowered to detect potentially important differences.
Conclusion
Medial clavicle fractures have favorable functional outcomes and pain relief at minimum 1-year follow-up among those patients who survive the trauma, but a high proportion will die within 3 years of the injury. This likely reflects both the frailty of a predominantly older patient population and the fact that these often are high-energy injuries. The outcome measures in our cohort were not associated with fracture displacement, shortening, or comminution; however, our sample size was underpowered on these points, and so these findings should be considered preliminary. Further studies are needed to determine the subset of patients with this injury who would benefit from surgical intervention.
Level of Evidence
Level IV, therapeutic study.
Introduction
Medial-third clavicle fractures are uncommon injuries that represent 2% to 15% of all clavicle fractures [3, 22, 24, 27]. In adults, medial clavicle fractures most commonly occur in men in the third to fifth decades of life [3, 22, 24, 27]. These injuries have been associated with high-energy injury mechanisms and multisystem trauma, most commonly involving the thorax [4, 27]. Most reported medial clavicle fractures have been treated nonoperatively, with the likelihood of union exceeding 95% [3, 22]. However, there is also interest in operative management of these injuries, as Brinker et al. [8] reported that up to 14% of patients developed nonunions in displaced medial clavicle fractures after nonoperative treatment. Bakir et al. [4] reported the results of 19 medial clavicle fractures, 31.6% (6 of 19) of which were treated surgically for displacement, although there was no reported follow-up for these patients.
The current evidence base on this rare injury has focused on union rates, demographics, and injury mechanism [3-6, 8, 9, 13, 14, 18, 22, 26]. The treatment strategies, although varied, have favored nonoperative management; however, prior studies are limited in the reporting of patient-reported outcome measures after these injuries and have not correlated fracture characteristics with outcome measures [7, 8, 9, 12-14, 16-21, 26].
The objectives of this study therefore were (1) to determine minimum 1-year functional outcomes using QuickDASH scores and pain scores after medial clavicle fractures and (2) to identify factors associated with these outcome variables.
Patients and Methods
Study Design and Setting
We retrospectively evaluated the records of all patients with medial clavicle fractures treated at two tertiary Level I trauma centers (Brigham and Women’s Hospital and Massachusetts General Hospital) in one metropolitan area from January 2010 to March 2019. We queried our billing records database using the Common Procedural Terminology codes 23500 (closed treatment of clavicular fracture; without manipulation) and 23515 (open treatment of clavicular fracture, includes internal fixation, when performed) for patients treated nonoperatively and surgically, respectively, during the study period.
Patient Selection
The medical records and radiographs of 1316 patients who underwent nonoperative treatment and 634 patients who underwent surgical treatment of clavicle fractures between January 2010 and March 2019 were screened, and 74 adult (older than 18 years at the time of injury) patients with medial-third clavicle fractures were identified as potentially eligible. Medial clavicle fractures were defined according to the Craig modification of the Allman classification (fracture in the proximal one-third of the clavicle) [1]. Patients were excluded if they were younger than 18 years of age, had a segmental clavicle fracture, pathologic fracture, or had less than 1 year of follow-up postinjury. This included 72 patients treated nonoperatively and two patients treated surgically. Nonoperative treatment was adjusted at the discretion of the treating surgeon, with common adjuncts being a sling for comfort, physical therapy, and early ROM. Of the included patients, one patient treated surgically and one patient treated nonoperatively were excluded from the study for clinical follow-up less than 1 year. Seventy-four adult patients with 74 medial clavicle fractures were contacted and their histories reviewed. Twenty-six patients were deceased, leaving 48 patients. An additional three patients declined to participate, and 14 patients could not be contacted. Sixty-five percent (31 of 48) of patients completed follow-up and were analyzed as our study group, with 94% (29 of 31) treated nonoperatively and 7% (2 of 31) treated surgically (Fig. 1). Surgery was indicated for one patient due to proximity of the medial fragment to the superior vena cava, while the other was a high-level athlete who was fixed for return to sport.
Fig. 1.
Study inclusion flow diagram.
Characteristics of the Study Group
The mean age of the 31 patients in our study was 65 ± 20 years at the time of injury, the mean American Society of Anesthesiologists score was 2 ± 1, and the mean Charlson Comorbidity Index score was 3 ± 2, representing a relatively older patient group with medical comorbidities. Sixty-five percent (20 of 31) of the patients were men, and the mean BMI was 25 ± 4 kg/m2. All patients in our cohort were white and English-speaking, and the mean Area Deprivation Index was in the 23rd ± 21st national percentile, representing an affluent group. The primary health insurance was nearly evenly split between private insurance (52% [16 of 31]) and Medicare (48% [15 of 31]). The dominant upper extremity was injured in 42% (13 of 31) of patients. Sixty-one percent (19 of 31) of injured patients were employed, and 13% (4 of 31) had a manual labor occupation. Seventy-one percent (22 of 31) of the patients had a high-energy mechanism of injury, and 71% (22 of 31) presented with a nonisolated injury (Table 1).
Table 1.
Characteristics of patients with medial clavicle fractures (n = 31)
| Characteristic | Data |
| Age in years | 65 ± 20 |
| BMI in kg/m2 | 25 ± 4 |
| Area Deprivation Index (United States percentile) | 23 ± 21 |
| Displacement in mm | 7 ± 7 |
| Shortening in mm | 5 ± 6 |
| ASA classification | 2 ± 1 |
| CCI classification | 3 ± 2 |
| Gender, men | 65 (20) |
| English-language speaker | 100 (31) |
| Race, white, as self-reported | 100 (31) |
| Insurance | |
| Private | 52 (16) |
| Medicare | 48 (15) |
| Employed | 61 (19) |
| Manual labor occupation | 13 (4) |
| Dominant upper extremity | 42 (13) |
| Diabetes mellitus | 6 (2) |
| Smokes cigarettes currently | 3 (1) |
| Nonisolated injury | 71 (22) |
| High-energy mechanism | 71 (22) |
| Nondisplaced fracture | 35 (11) |
| Fracture comminution | 39 (12) |
| Surgical treatment | 6 (2) |
Data presented as mean ± SD or % (n); ASA = American Society of Anesthesiologists; CCI = Charlson Comorbidity Index.
At the time of injury, 16% (5 of 31) of fractures were imaged with plain radiography alone, whereas 84% (26 of 31) of fractures were imaged with both plain radiography and CT. The indication for CT was nonstandardized and in many instances was part of a polytrauma evaluation protocol. Thirty-five percent (11 of 31) of fractures were nondisplaced; 39% (12 of 31) had fracture comminution. The mean fracture displacement was 7 ± 7 mm, and the mean fracture shortening was 5 ± 6 mm (Table 1). According to the Edinburgh classification of medial clavicle fractures [23], six fractures were Type 1A1, six were Type 1A2, 14 were Type 1B1, and five were Type 1B2 (Fig. 2).
Fig. 2.
The Edinburgh classification for medial clavicle fractures (Type 1) is shown. Type 1A1 fractures are extraarticular, nondisplaced fractures; Type 1A2 are extraarticular, displaced fractures; Type 1B1 are intraarticular, nondisplaced fractures; and Type 1B2 are intraarticular, displaced fractures.
The 31 patients with telephone follow-up were followed clinically by the treating provider for a mean (range) of 3 ± 3 months (0 to 11). Sixty-eight percent (21 of 31) of patients had radiographic follow-up at least 6 weeks after surgery, and radiographic union was achieved in 19 of 21 patients at a mean of 13 ± 7 weeks (6 to 30). One patient who achieved radiographic union had persistent pain and was not classified as clinical nonunion. There were no patients with clinical nonunion. Of these 21 patients, two were treated with surgery and 19 nonoperatively. There were no patients with nonunion among the surgical patients and two patients with radiographic nonunion in the nonsurgical group. The patients with radiographic nonunion reported pain scores of 0 at the time of follow-up. The remaining patients had insufficient radiographic follow-up and clinical documentation to assess fracture healing, with a mean radiographic follow-up of 13 ± 9 days.
Data Sources and Analysis
The electronic medical records of the cohort, including imaging studies, were reviewed by a physician investigator not involved with treatment or diagnosis (MHL, PG, LL) to assess demographic characteristics, fracture characteristics, and fracture union. Radiographic union was defined on plain radiography by the presence of a bridging callus on two or more views or on CT by the presence of a bridging callus across the fracture site. Radiographic delayed union was characterized as failure to reach radiographic union at 6 months, and nonunion was classified as the failure to reach radiographic union at 9 months. Clinical union was determined through record review; the patient’s fracture was considered clinically united if the patient had no pain, a normal weightbearing status, and no fracture mobility on examination. Clinical delayed union was defined at 6 months and nonunion at 9 months. If either radiographic or clinical union could not be determined via record review, the union type was considered unknown. We attempted to call each patient over the telephone to collect self-reported functional outcome measures and pain scores at a minimum of 1 year after injury. To fully verify living and deceased status, all patients were searched with the most recent Social Security Death Index and public obituaries based on the location of last residence [29]. If death could not be confirmed, we conducted an additional electronic medical record search, then we contacted the people listed in the EMR to verify the deceased status. Confirmation of deceased status was obtained for all 74 patients.
Response Variable and Explanatory Variables
Our primary response variable was the QuickDASH score at the time of telephone follow-up. Our secondary response variable was the pain score as rated on a 0-point to 10-point Likert scale. Follow-up for self-reported functional outcomes and pain were obtained at a mean (range) of 5 years (2 to 10) from the time of injury. We studied the following explanatory variables: age, gender, race, dominant hand injury, employment status, manual labor occupation, primary health insurance, social deprivation, BMI, diabetes mellitus, smoking status, American Society of Anesthesiologists physical status classification, Charlson Comorbidity Index, nonisolated injury, high-energy mechanism of injury, nondisplaced fracture, fracture comminution, superior-inferior fracture displacement, medial-lateral fracture shortening, and surgical treatment of the medial clavicle fracture. Social deprivation was measured using the Area Deprivation Index, a validated tool for assessing social deprivation that accounts for poverty, housing, employment, and education using patient-reported address of residence [30]. The Area Deprivation Index is expressed as a United States national percentile in this study, in which higher percentiles represented more social deprivation. The BMI closest to the date of injury (within 1 year before or after treatment) was used for analysis. High-energy injuries in this study comprised motor vehicle collisions and falls from height. Fracture comminution, displacement, and shortening were measured on injury frontal plain radiographs showing the greatest displacement, using the method described by Hill et al. [13].
Ethical Approval
Ethical approval for this study was obtained from Partners Healthcare, Somerville, MA, USA (approval number 2010P002562/PHS).
Statistical Analysis
We calculated descriptive statistics for explanatory variables for the study cohort. Nearly all explanatory variables had 100% data completeness, except for BMI, which had 94% data completeness. All variables were analyzed using the data available, and missing data were excluded. We performed linear regression analysis to identify continuous variables associated with our outcome measures, and we performed the Mann-Whitney U test to identify dichotomous variables associated with our outcome measures.
We used a convenience sample. Because multiple comparisons were performed, we did not use the standard significance criterion of α = 0.05; we performed a Bonferroni correction, and we considered an α < 0.0025 as statistically significant for all statistical tests. The standard power criterion of (1-β) = 0.80 was also used.
Results
Functional Outcome (QuickDASH) and Pain Score
The mean QuickDASH score was 12 ± 15, and the mean pain score was 1 ± 1. At the time of telephone follow-up, 35% (26 of 74) of the eligible patients were deceased.
The 1-, 3-, and 5-year mortality rates were 15% (11 of 74), 22% (16 of 74), 34% (25 of 74), respectively. Of those who died in the first year after injury, three died in the first 30 days and seven died within the first 90 days, signifying a 4% 1-month and a 9% 90-day mortality rate. Nine of the total deceased patients had experienced high-energy trauma and 17 from low-energy mechanisms such as ground level falls.
Factors Associated with QuickDASH and Pain Scores
No factors were associated with the QuickDASH score or pain score. Continuous variables, including age, fracture displacement, and fracture shortening, were analyzed using linear regression. Increasing age (beta regression coefficient 0.17 [95% confidence interval -0.12 to 0.45]; p = 0.25), increasing fracture displacement (beta regression coefficient -0.48 [95% CI -1.34 to 0.39]; p = 0.27), and increasing fracture shortening (beta regression coefficient -0.39 [95% CI -1.42 to 0.64]; p = 0.44) were not associated with QuickDASH score. Dichotomous categorical variables, including gender and surgical treatment, were analyzed using the Mann-Whitney U test. Male gender (median 3.41 [interquartile range 0 to 18.19] versus 2.30 [IQR 0 to 28.75], Hodges-Lehmann median difference 0 [95% CI -4.55 to 14.32]; p = 0.49) and surgical treatment (median 2.30 [IQR 0 to 20.00] versus 3.41 [IQR 2.84 to 3.98], Hodges-Lehmann median difference -0.03 [95% CI -34.13 to 4.55]; p = 0.97) were not associated with QuickDASH score. Male gender (median 0 [IQR 0 to 0.25] versus 0 [IQR 0 to 0.5], Hodges-Lehmann median difference 0 [95% CI 0 to 0]; p = 0.89) and surgical treatment (median 0 [IQR 0 to 0] versus 0 [IQR 0 to 1], Hodges-Lehmann median difference 0 [95% CI -2 to 0]; p = 0.43) were not associated with pain score.
A post-hoc power analysis was performed. Assuming a mean QuickDASH score of 10 and SD of 15, a sample size of 31 had 80% power to detect an approximately 20-point increase in the QuickDASH score in the bivariate analysis, assuming equally sized, dichotomous groups. We were underpowered to detect less than a 20-point difference in our primary outcome variable, the QuickDASH score, with our sample size.
Discussion
Fractures of the medial clavicle constitute a rare but unique subset of clavicle fractures. Occurring with an estimated incidence of 0.82 per 100,000 people per year [22], they account for approximately 2% to 15% of all clavicle fractures [22, 24, 27]. Medial clavicle fractures are associated with high-energy trauma [21, 27] and commonly occur in patients with multiple injuries [22]. Nonoperative treatment has been reported to have low rates of nonunion and pain and low rates of secondary intervention [2], and the indications for surgical fixation are unclear [4, 21]. Asadollahi and Bucknill [3] in 2019 performed a systematic review of research on medial clavicle fractures, and only the research of Salipas et al. [24] assessed validated patient-reported outcome measures in a series of more than 10 patients. Throckmorton and Kuhn [27] published a series of 55 patients with medial clavicle fractures, but reported outcomes were a range of “no pain” to “severe pain.” Our research helps to address the paucity of reporting on validated patient-reported outcome measures, patient-reported pain scores, and mortality after medial clavicle fractures; moreover, attempts have not been made to correlate fracture characteristics with outcome measures to guide the decision for surgical treatment. We found that the minimum 1-year functional outcomes and pain scores after medial clavicle fractures are favorable among those patients who survive the trauma, but a high proportion will die within 3 years of the injury. We were underpowered to detect factors associated with functional outcomes after this injury.
Limitations
This study has several limitations. First, a selection bias may have occurred due to nonstandardized treatment approaches. Treating surgeons may have selected more displaced fractures for operative treatment, which may confound the functional outcomes and decrease our ability to detect a difference in functional outcomes based on fracture characteristics. Similarly, the nonoperative treatment and rehabilitation protocol was nonstandardized, which may have influenced the results in ways that are difficult to characterize. Second, the substantial mortality rate may have contributed to survivorship bias by selecting for better patient-reported outcomes if frailer and more severely traumatized patients were deceased at the time of the outcome collection. Thirdly, assessment bias may have occurred in the method of data collection. Although validated metrics were used, telephone interviews do not allow for physical examinations and are therefore a limited evaluation of functional outcomes. Fourth, telephone interviews can be subject to recall bias; however, we used validated outcome measures that focused on function and pain at the time of the telephone interview rather than at time of injury. Fifth, our study may be affected by transfer bias if a minimum 1-year follow-up is insufficient to assess outcomes such as posttraumatic arthritis. It may be the case that further functional decrements may arise if outcomes were reassessed with longer term follow-up. Sixth, the response rate of 65% (31 of 48) leaves a portion of living patients out of the study and limits the power of the investigation. Seventh, treatment was not standardized and at the discretion of the treating surgeon. The most common additional treatment was physical therapy, which was used in 29% (9 of 31) of patients. Further investigations are warranted to understand which treatments help improve results. Eighth, our sample size was underpowered to identify factors associated with the outcomes of medial clavicle fractures. Finally, this study was conducted at two Level I trauma centers, which may limit the generalizability of our conclusions.
Functional Outcome (QuickDASH) and Pain Score
We found the 5-year mortality rate was 35%, which exceeds the expected mortality rate of the United States population of 9.2% in this age group [2, 31]. The higher-than-expected mortality rate after this injury is likely multifactorial. It may underscore the underlying frailty of this population as some injuries resulted from innocuous mechanisms such as a ground-level fall, despite the robust anatomic restraints of the medial clavicle [10, 11, 15]. This is reflected in the high proportion of deceased patients who suffered low-energy injury (17) compared with those with a high-energy mechanism (9). Alternatively, it may be reflective of the severity-of-injury burden that is reflected in the elevated 30- and 90-day mortality (4% and 9%, respectively) of this cohort.
Setting aside the notable mortality rate after this injury, nonoperative treatment of medial clavicle fractures was associated with acceptable QuickDASH and pain relief in the cohort of patients available for follow-up. Our functional findings among the surviving patients support those of previous reports showing that medial clavicle fractures have favorable outcomes with nonoperative treatment [3, 22]. Asadollahi and Bucknill [3] systematically reviewed seven case series and 10 case reports of acute medial clavicle fractures and found similar rates of union between surgical and nonoperative treatment. The authors concluded that favorable functional outcomes and predictable union can be expected from the nonoperative treatment of these injuries. Our findings should be interpreted in the context of the limitations of our study. The mean (range) follow-up after injury was 5 years (1 to 10) but with minimum 1-year follow-up after injury, and it is possible that the QuickDASH score and pain score may evolve or deteriorate with longer follow-up. We had a high proportion of patients who were lost to follow-up, and it may be that patients with worse QuickDASH scores and pain scores were more likely to be unavailable for review.
Factors Associated with QuickDASH and Pain Scores
In this study, we did not identify factors associated with functional outcome or pain after medial clavicle fractures; however, we were underpowered for this analysis because of the rarity of this injury and limited sample size. Previous studies suggested that fully displaced fractures may be associated with higher risk of nonunion but were underpowered to detect differences, and incidence varied depending on classification [23]. Historically, fracture displacement, fracture comminution [25], and periarticular location [28] have been cited as indications for surgical treatment of these injuries. In our series, we did not detect correlation between fracture parameters and patient-reported disability or pain. Since our study was underpowered, we cannot recommend for or against the use of radiographic fracture parameters to indicate surgical treatment of medial clavicle fractures. We encourage future studies on clinical and radiographic factors associated with functional outcomes and pain after this uncommon injury.
Other Relevant Findings
The demographics of our study cohort are comparable to those in published reports. The predominance of this injury in men is supported by earlier studies [3, 5-6, 22, 27]. We found medial clavicle fractures to be 4.1% of all clavicle fractures treated in the time included, in keeping with the work of others [22, 23, 27]. Of the cohort with sufficient radiographic follow-up, there was a 9.5% (2 of 21) incidence of radiographic nonunion, with two occurring in conservatively treated patients and none in surgically treated patients. This incidence is notable and potentially higher than previous studies; however, this finding is underpowered, may represent Type II error, and may include possible selection bias for those who had persistent symptoms requiring longer follow up [2-4, 10]. Both patients with radiographic nonunion were asymptomatic at last clinical follow-up. Moreover, 71% of the fractures in the cohort were associated with high-energy trauma, such as traffic collisions and falls from height, a finding supported by Asadollahi and Bucknill [3]. Throckmorton and Kuhn [27] and Robinson [22] also found that between 54% and 88% of these injuries were associated with similar high-energy mechanisms. Oe et al. [21] reported high-energy injury mechanisms in 100% of patients; however, their case series of 10 patients with displaced medial clavicle fractures treated operatively may have selected patients with high-energy mechanisms. Anatomically, the medial end of the clavicle is enveloped by a robust joint capsule, and the sternoclavicular ligaments confer greater resistance to torsion and bending than the midshaft or distal clavicle [15]; therefore, higher energy is needed to break these attachments, thus making the medial clavicle fracture a less commonly observed injury. The high proportion of concomitant injuries with medial clavicle fractures is supported by previous studies citing proportions as high as 90% [3, 6, 27].
Conclusion
Fractures of the medial third of the clavicle are uncommon. In our population, derived from two urban tertiary trauma centers, these injuries were seen more frequently in men in or beyond the seventh decade of life. They were associated with high-energy mechanisms and usually copresented with other traumatic injuries. Thirty-five percent (26 of 74) of patients were deceased at a mean of 2 years after injury, likely related to the frailty of these patients and/or the severity of associated injuries. QuickDASH and pain scores were favorable in the surviving available cohort, suggesting adequate patient-reported functional and pain outcomes; however, our conclusions should be interpreted in light of the high proportion of patients who were lost to follow-up. We cannot exclude the possibility that patients with poorer functional outcomes and problems with fracture healing may have been unavailable for our review. We were unable to detect factors associated with QuickDASH and pain scores, but our study was underpowered to detect differences due to the sample size of this uncommon injury. We did not find evidence to support operative treatment of fractures with worse initial radiographic parameters. Treating physicians could consider long-term radiographic and clinical follow-up to evaluate functional deficits or disability over time. Future studies could include correlations between medial clavicle fractures and bone density or longitudinal studies of functional outcomes. Future multicenter, prospective studies of medial clavicle fractures may generate appropriate power to identify factors associated with functional outcomes and pain after this injury and further elucidate operative indications.
Footnotes
Each author certifies that there are no funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Ethical approval for this study was obtained from Partners Healthcare, Somerville, MA, USA (approval number 2010P002562/PHS).
This work was performed at Brigham and Women’s Hospital and Massachusetts General Hospital.
Contributor Information
Phillip Grisdela, Email: pgrisdela@bwh.harvard.edu.
Laura Lu, Email: llu15@mgh.harvard.edu.
Dafang Zhang, Email: dzhang9@bwh.partners.edu.
Brandon Earp, Email: Bearp@bwh.harvard.edu.
References
- 1.Allman FL. Fractures and ligamentous injuries of the clavicle and its articulation. J Bone Joint Surg Am. 1967;49:774-784. [PubMed] [Google Scholar]
- 2.Arias E, Xu J. United States life tables, 2017. National Health Statistics Reports. 2019;68:1-66. [PubMed] [Google Scholar]
- 3.Asadollahi S, Bucknill A. Acute medial clavicle fracture in adults: a systematic review of demographics, clinical features and treatment outcomes in 220 patients. J Orthop Traumatol. 2019;20:24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bakir MS, Merschin D, Unterkofler J, et al. Injuries of the medial clavicle: a cohort analysis in a level-I trauma center. Concomitant injuries. Management. Classification. Chirurgia (Bucur). 2017;112:594. [DOI] [PubMed] [Google Scholar]
- 5.Bakir MS, Unterkofler J, Haralambiev L, et al. Medial injuries of the clavicle: more prevalent than expected? A big data analysis of incidence, age, and gender distribution based on nationwide routine data. Eur J Trauma Emerg Surg. Published online January 20, 2020. DOI: 10.1007/s00068-019-01293-0. [DOI] [PMC free article] [PubMed]
- 6.Bakir MS, Unterkofler J, Hönning A, et al. Shoulder girdle injuries involving the medial clavicle differ from lateral clavicle injuries with a focus on concomitant injuries and management strategies: a retrospective study based on nationwide routine data. PLoS One. 2019;14:e0224370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bartonícek J, Fric V, Pacovský V. Displaced fractures of the medial end of the clavicle: report of five cases. J Orthop Trauma. 2010;24:e31-5. [DOI] [PubMed] [Google Scholar]
- 8.Brinker MR, Edwards TB, O'Connor DP. Estimating the risk of nonunion following nonoperative treatment of a clavicular fracture. J Bone Joint Surg Am. 2005;87:676-677. [DOI] [PubMed] [Google Scholar]
- 9.Chen J, Xu D, Luo P. A new technique for medial-end comminuted clavicle fractures. Injury. 2019;50:1261. [DOI] [PubMed] [Google Scholar]
- 10.Clement ND, Aitken SA, Duckworth AD, McQueen MM, Court-Brown CM. The outcome of fractures in very elderly patients. J Bone Joint Surg Br. 2011;93:806-810. [DOI] [PubMed] [Google Scholar]
- 11.Clement ND, Aitken S, Duckworth AD, McQueen MM, Court-Brown CM. Multiple fractures in the elderly. J Bone Joint Surg Br. 2012;94:231-236. [DOI] [PubMed] [Google Scholar]
- 12.Fransen P, Bourgeois S, Rommens J. Kirschner wire migration causing spinal cord injury one year after internal fixation of a clavicle fracture. Acta Orthop Belg. 2007;73:390-392. [PubMed] [Google Scholar]
- 13.Hill JM, McGuire MH, Crosby LA. Closed treatment of displaced middle-third fractures of the clavicle gives poor results. J Bone Joint Surg Br. 1997;79:537-539. [DOI] [PubMed] [Google Scholar]
- 14.Kim KC, Shin HD, Cha SM. Surgical treatment of displaced medial clavicle fractures using a small T-shaped plate and tension band sutures. Arch Orthop Trauma Surg. 2011;131:1673-1676. [DOI] [PubMed] [Google Scholar]
- 15.Lee JT, Campbell KJ, Michalski MP, et al. Surgical anatomy of the sternoclavicular joint: a qualitative and quantitative anatomical study. J Bone Joint Surg Am. 2014;96:e166. [DOI] [PubMed] [Google Scholar]
- 16.Liu H, Peng C, Zhang Z, et al. Single-center experience in the treatment of extremely medial clavicle fractures with vertical fixation of double-plate: a retrospective study. Medicine (Baltimore). 2020;99:e19605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Liu Z, Zhang J, Tian X, Kan S. Displaced medial-end clavicle fractures treated with locking plate osteosynthesis. Med Sci Monit. 2019;25:7591-7596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Low AK, Duckworth DG, Bokor DJ. Operative outcome of displaced medial-end clavicle fractures in adults. J Shoulder Elbow Surg. 2008;17:751-754. [DOI] [PubMed] [Google Scholar]
- 19.Lyons FA, Rockwood CA. Migration of pins used in operations on the shoulder. J Bone Joint Surg Am. 1990;72:1262-1267. [PubMed] [Google Scholar]
- 20.Nakayama M, Gika M, Fukuda H, et al. Migration of a Kirschner wire from the clavicle into the intrathoracic trachea. Ann Thorac Surg. 2009;88:653-654. [DOI] [PubMed] [Google Scholar]
- 21.Oe K, Gaul L, Hierholzer C, et al. Operative management of periarticular medial clavicle fractures-report of 10 cases. J Trauma Acute Care Surg. 2012;72:E1-E7. [DOI] [PubMed] [Google Scholar]
- 22.Robinson CM. Fractures of the clavicle in the adult. Epidemiology and classification. J Bone Joint Surg Br. 1998;80:476‐484. [DOI] [PubMed] [Google Scholar]
- 23.Robinson CM, Court-Brown CM, McQueen Margaret M, Wakefield AE. Estimating the risk of nonunion following nonoperative treatment of a clavicular fracture. J Bone Joint Surg Am. 2004;86:1359-1365. [DOI] [PubMed] [Google Scholar]
- 24.Salipas A, Kimmel LA, Edwards ER, Rakhra S, Moaveni AK. Natural history of medial clavicle fractures. Injury . 2016;47:2235-2239. [DOI] [PubMed] [Google Scholar]
- 25.Sidhu VS, Hermans D, Duckworth DG. The operative outcomes of displaced medial-end clavicle fractures. J Shoulder Elbow Surg. 2015;24:1728-1734. [DOI] [PubMed] [Google Scholar]
- 26.Siebenlist S, Sandmann G, Kirchhoff C, Biberthaler P, Neumaier M. Anatomically precontoured LCP for delayed union of a medial third clavicle fracture. Case report with review of the literature. Acta Chir Orthop Traumatol Cech. 2013;80:407-410. [PubMed] [Google Scholar]
- 27.Throckmorton T, Kuhn JE. Fractures of the medial end of the clavicle. J Shoulder Elbow Surg. 2007;16:49‐54. [DOI] [PubMed] [Google Scholar]
- 28.Titchener A, See A, Van Rensburg L, Tytherleigh-Strong G. Displaced medial end clavicular fractures treated with an inverted distal clavicle plate contoured through 90 degrees. J Shoulder Elbow Surg. 2019;28:e97-e103. [DOI] [PubMed] [Google Scholar]
- 29.United States Social Security Death Index. Database Search. Avaliable at https://www.familysearch.org/search/collection/1202535. Accessed September 20, 2020.
- 30.Wright MA, Beleckas CM, Calfee RP. Mental and physical health disparities in patients with carpal tunnel syndrome living with high levels of social deprivation. J Hand Surg Am. 2019;44:335.e1-335.e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Xu J, Sherry LM, Kochanek KD, et al. Mortality in the United States, 2018. NCHS Data Brief. 2020;355:1-8. [PubMed] [Google Scholar]