Fig. 6.1
Case 1, Upright anteroposterior (a) and 45° cephalic tilt (b) views showing a midshaft clavicle fracture
Patients should have upright radiographs obtained within the first 2 weeks. Even nondisplaced clavicle fractures can show significant instability and displacement on subsequent X-rays over 25% of the time, occasionally changing treatment decisions [27]. Higher energy mechanisms of injury are associated with increased likelihood of early displacement of initially minimally displaced fractures [28]. After the first 2 weeks, AP and cephalad tilt X-rays are obtained at 4 to 6 week intervals until radiographic union is obtained.
Indications
Today, widely accepted indications for nonoperative management in adults are nondisplaced or minimally displaced midshaft clavicle fractures and patients who are poor surgical candidates based on medical comorbidities . While open fractures or threatened skin compromise are indications for operative treatment, there are differing opinions regarding the optimal care of clavicle fractures that do not fall into one of the two aforementioned indications. Relative indications for nonsurgical management of clavicle fractures include displacement <1.5 cm, no comminution, bony contact, and shortening of less than 1.5–2 cm. Patients who do not meet these parameters have a higher probability of satisfactory functional and pain outcomes with surgical management [29–31]. Patient activity level, age, gender, soft tissue envelope, handedness, smoking habits, fracture pattern, and mechanism of injury are additional factors that impact management.
Robinson et al. have shown that female gender, increasing age, displacement, and comminution are all predictors of nonunion in midshaft clavicle fractures treated without surgery [29]. Patients should be counseled regarding the importance of smoking cessation and should be encouraged, at least during the period of healing to stop or decrease their smoking [32, 33]. Patients who smoke are four times more likely to progress to nonunion when midshaft clavicle fractures are treated nonsurgically [32, 33].
Murray et al. evaluated risk factors for developing nonunions in displaced midshaft clavicle fractures. They created a “Ready Reckoner” to allow practitioners to estimate the risk of nonunion with nonoperative treatment. They evaluated 941 adult patients with closed displaced midshaft clavicle fractures. Their predictor table is able to provide an expected risk of nonunion for displaced midshaft clavicle fractures treated nonoperatively. These factors included the overall radiographic displacement, smoking history, and fracture comminution. The overall expected risk of nonunion increases with smoking, comminution, and increasing displacement. Operative management of all displaced clavicle fractures requires 7.5 procedures to prevent a single nonunion according to the study. However, using the statistical information from the study, this result may be significantly improved. If only those patients with an estimated 40% or greater risk of nonunion were treated surgically, the number of procedures required to prevent a single nonunion dropped to 1.7 operations [33].
Goals of Treatment
The goal of treatment is not only radiographic union but, most importantly, returning the patient’s shoulder back to as close to preinjury functional level as possible. Neer asserted in Rockwood and Green’s Fractures, in adults “a certain amount of deformity is to be expected, [but] generally [is] compatible with satisfactory return of function in the shoulder” and even completely displaced fractures “generally do well with nonoperative management” [34]. Much of the early literature focused on radiographic healing as the primary outcome, but over the last two to three decades, there has been increased focus on the importance of more patient reported outcomes in the literature.
Treatment Options
Immobilization of clavicle fractures is generally via figure of eight brace or simple shoulder sling, with the latter being more commonly utilized. A prospective, randomized study showed similar fracture healing between these two treatment methods, but showed that patients had 26% dissatisfaction with the figure of eight brace treatment protocol at 3 months compared to 7% dissatisfaction in the cohort treated with a sling [35]. In a more recent prospective, randomized study, the figure of eight brace also caused significantly more skin complications, most often in the axilla, than the sling. The figure of eight brace has also been suggested to be more painful during the first treatment day than the sling and more difficult for patients and their families to apply than a sling [36]. These have led to the decreased use of the figure of eight brace. The final fracture union with either method of immobilization showed minimal change from the initial amount of displacement, substantiating Neer’s statement that there is no good way to hold an attempted reduction of clavicle fractures [35].
Treatment Course
Nonoperative management generally entails a period of 2–6 weeks of immobilization of the shoulder girdle for patient comfort with restricted overhead activities. The arm is generally held in internal rotation with the elbow flexed to approximately 90° in a sling [36]. Patients should be encouraged to remove the sling at several points during the day to work on elbow range of motion to prevent elbow stiffness [36]. Immobilization can be discontinued and light activities below shoulder level can be resumed as patient comfort allows. Overhead activities are generally restricted for the first 6 weeks. By limiting the abduction and forward flexion to 90° the rotation of the clavicle is limited, minimizing rotational and torsional forces on the acutely healing clavicle. At approximately 6 weeks, most patients are ready to progress to activities as tolerated.
Nonoperative Treatment Adjuncts
In an effort to expedite healing or for treatment of delayed or nonunions, low-intensity pulsed ultrasound (LIPUS ) has been attempted. When looking at all fractures the effect of LIPUS is inconsistent, with some randomized studies showing benefit and others showing no benefit [37–40]. A large Cochrane systematic review of 11 heterogeneous randomized or quasi-randomized controlled trials evaluating LIPUS in treating various acute fractures found equivocal results regarding, function, time to union, or pain. They suggest that there is insufficient evidence for the routine use of LIPUS in acute fractures [41]. In contrast, a recent meta-analysis evaluating healing in all bones, not just clavicle fractures, suggested improved outcomes at 6 months in acute fractures treated with LIPUS [42]. Once nonunion has developed, LIPUS showed improved healing in numerous long bone fractures, but again the clavicle was not specifically studied and only made up a small proportion of the included cases (1 out of 29) [43]. There is only one randomized controlled trial evaluating LIPUS in acute midshaft clavicle fractures. A level I study in which 101 adult patients with isolated midshaft clavicle fractures were randomized to either LIPUS or placebo ultrasound probe. They found no difference between placebo and LIPUS in regards to pain, pain medication requirement, return to activities, or subjective healing in the treatment of acute midshaft clavicle fractures. Future level 1 randomized trials are necessary to further evaluate the suggested benefits of LIPUS specific to midshaft clavicle fractures.
Outcomes
Despite the recent trend to surgically fix more midshaft clavicle fractures, authors caution against routine ORIF for all midshaft clavicle fractures [44–47]. Nordqvist et al. retrospectively reviewed outcomes of 225 patients with midshaft clavicle fractures treated nonoperatively with 197 of these placed into a figure of eight brace for approximately 3 weeks. Of the included fractures, 32% of the fractures included were either minimally or nondisplaced. Of all patients, 185 had “good” outcomes and reported an asymptomatic shoulder where 39 had moderate pain and were rated “fair” and only 1 rated “poor”. There were 53 malunions, of which 40 were asymptomatic. There were seven nonunions, of which three were asymptomatic. They suggest that surgery should not be considered on an acute closed midshaft clavicle fracture without neurovascular or skin compromise [48]. A prospective study of 60 patients treated nonoperatively showed that regardless of the method of nonoperative treatment (sling vs. figure of eight), there was 100% radiographic union at 14 weeks with lesser than 1 cm shortening. Patients ultimately attained full function with Constant and American Shoulder and Elbow Surgeon (ASES) scores in the mid-90s on a 100-point scale with 100 being optimal. One caveat of this study is that 40% of these patients were minimally or nondisplaced [36]. Faldini et al. studied 100 displaced Edinburgh 2B midshaft clavicle fractures treated in a figure of eight brace and 81% had excellent results, 12% had good results, 5% had fair results, and only 2% had poor results based on the Disability of Arm Shoulder and Hand (DASH) scores with a 3% nonunion rate. They concluded that “nonoperative treatment is still appropriate in most cases, as it yields good results without incurring the potential complications of surgery” [49].
Complications
Patients should be aware of all risks and benefits of nonoperative treatment of clavicle fractures during initial patient evaluation. Patients should be counseled to expect approximately 4–12 weeks of pain and decreased function after injury.
Complications of both nonsurgical and surgical treatment of clavicle fractures include cosmetic deformity, nonunion, malunion, pain, local irritation, functional limitations of the shoulder, and need for future surgery. Cosmetic deformity from the “bump” alone has been reported to be the main source of patient dissatisfaction in approximately 40% of those patients who are ultimately unhappy with the outcome of nonoperative treatment [50].
When all complications are included, McKee et al. has shown a 42% complication rate for nonoperative treatment compared with 29% for surgical management of displaced clavicle fractures. However, the most common complications, excluding cosmetic deformity, for nonoperative treatment were nonunion, symptomatic malunion, and neurologic difficulty, whereas the most common operative complications were hardware irritation, pin protrusion, and wound infection [46]. However, it is cautioned that their results do not suggest surgery is necessary for all fractures as this would unnecessarily expose many patients, which would have healed well without surgery, to the risks associated with surgical treatment [47]. Nonoperative management of clavicle fractures avoids the complications specific to surgical treatment especially infection, symptomatic hardware, and anterior chest wall numbness.
Nonunions and malunions are widely accepted as the most frequent complications associated with midshaft clavicle fractures. Patients can have pain, neurologic changes, shoulder dysfunction, and clavicle deformity associated with improper union.
The definition of a clavicle nonunion has not been consistently defined, but is often absence of radiographical healing after 4–6 months [12, 29]. Research from the 1960s and 1970s demonstrated nonunion rates of only 0.1–0.8% for nonsurgical management. More recent literature shows a 0.5–5.9% nonunion rate for nonsurgical management of all midshaft clavicle fractures. However, the nonunion rate is significantly higher at up to 15.1% in completely displaced fractures treated nonoperatively [48, 51]. In the event of nonunion, delayed surgical treatment of nonunion with various techniques has shown 92–100% rates of successful union [52–55].
The majority of clavicle fractures treated without surgery will heal with some degree of deformity, with the most common being shortening with inferior displacement of the lateral fragment [12]. Malunion may be associated with symptoms including loss of shoulder function, loss of endurance, neurologic symptoms, cosmetic deformity, weakness, and pain in the shoulder [56]. Malunion with shortening >1.5 cm may be associated with increased pain [30].
McKee et al. evaluated range of motion, strength, and endurance in midshaft clavicle fractures treated nonoperatively. They found no loss of motion in forward flexion, abduction, or external rotation when compared to the noninjured side. There was a 19% loss of strength in flexion, 18% loss of strength in abduction, 19% loss of strength in external rotation, and 15% loss of strength in internal rotation. There was a 25% loss of endurance in flexion, 33% loss of abduction endurance, 18% loss of external rotation endurance, and 22% loss of internal rotation endurance. DASH and Constant scores showed a significant level of dysfunction at mean follow up of 55 months when compared to the general population [50]. Additionally, a study of midshaft clavicle fractures with final shortening of 15 mm or greater showed reduced peak abduction velocity [57]. However, nonoperative treatment of midshaft clavicle fractures is a viable option given that nonunion and malunion are both manageable complications of this treatment course.
Cost
The cost of treating midshaft clavicle fractures has increased relevance due to the recent emphasis on bundled payments and health care spending. Walton et al. compared the cost of surgical and nonsurgical treatment using 2013 Medicare values. The mean cost for surgical treatment was $14,763 and nonsurgical cost was $3113 with nonsurgical treatment saving $11,650 per patient. They used a set endpoint of either delayed surgery for nonoperative patients or reoperation in operative patients. Nonoperative management was more financially sound compared to initial operative treatment until the point that delayed surgery was required in 95% of nonoperative patients, and reoperation rates for patients treated with surgery initially were below 15%. In fact, this study suggests that treating all closed displaced midshaft clavicle fractures nonoperatively initially and then managing the complications of such when they develop is more fiscally responsible [45]. However, this study fails to assess the patient’s lost wages and decreased ability to contribute to society.
In contrast, a study of 204 patients accounting for lost wages, all hospital costs, and postoperative treatment found an ORIF treatment course cost $12,977 and a nonoperative treatment course cost $18,068 per patient. Operative patients had a mean lost earnings of $321 compared to $10,506 lost wages in the nonsurgically treated group. Additionally, operative patients required 3 days of assistance at home compared to 7 days of assistance for those treated nonoperatively, thus potentially causing lost wages in additional family members. Mean emergency room bills were similar, but the operative group did accrue an additional $8500 hospital bill. Operative patients utilized half the physical therapy of the nonoperative group ($900 vs. $1800) [58]. Other research shows that ORIF of clavicle fractures compared to nonoperative management is cost-effective only when the improved functional outcome associated with ORIF persists for 9 years [59].
It is unclear which treatment option is more fiscally responsible, likely due to the variability of surgical indications and the heterogeneity of these two study groups. The financial implications of the treatment of clavicle fractures will be an integral part of treatment, and likely topic of much discussion in the coming years.
Summary
The recent focus on functional outcomes and cost has called into question the historic nonoperative treatment of midshaft clavicle fractures. The modern widely accepted indications for nonoperative management of clavicle fractures include nondisplaced or minimally displaced fractures and those patients who may not be surgical candidates based on medical comorbidities. The two most commonly utilized nonoperative treatment options are a simple sling, or figure of eight brace, although the sling has become more popular in recent years. All patient factors including baseline activity level, handedness, fracture pattern, patient expectation, mechanism of injury, and risk factors for nonunion should be evaluated and discussed with the patient. Patients should be counseled that nonoperative management in most cases will provide a successful outcome, however some may require delayed surgery. A thorough discussion should occur with the patient including the risks and benefits of surgical and nonsurgical management of clavicle fractures to develop a treatment plan that is mutually acceptable to the patient and the physician.
Cases
Case 1
Case 1 is a 52-year-old right hand dominant male involved in a low-speed (10 mph) motorcycle collision. In the emergency room, the patient was found to have a small left pneumothorax, which did not require a chest tube. A left midshaft clavicle fracture was identified on his trauma chest radiograph. Upright clavicle radiographs were then obtained (Fig. 6.1a, b). He was admitted overnight to monitor the pneumothorax. He had a history of hypertension and never smoked. His left arm was placed in a sling at admission, and he was discharged from the hospital the next day after being cleared by general surgery. He was seen 8 days after his injury and his pain was well controlled on occasional ibuprofen. Radiographs showed minimal change from the emergency room upright films (Fig. 6.2a, b). He was counseled on treatment options including benefits and risks of both and opted for nonoperative treatment. At his 2-week visit, he reported that he was comfortable doing activities of daily living without his sling. We discussed treatment options again and he elected to continue nonoperative management. At his 6-week follow up, he was only having occasional mild fracture site pain. There was a palpable and visible bump over the midshaft clavicle at this visit (Fig. 6.3a, b). By his 12 week follow-up, he had returned to his job stocking shelves and occasionally lifting up to 50 lbs with only minimal pain. There was still a prominence over the midshaft clavicle. He started working on his golf swing around 4 months out from the injury and reported no discomfort. At the 8-month follow up visit, he reported no pain. Upright clavicle radiographs showed a z-deformity of the clavicle with incomplete union (Fig. 6.4a, b). Despite his incomplete radiographic healing, he was clinically completely asymptomatic, denied any functional deficits, and had full symmetric range of motion (Fig. 6.5a, b).
