The treatment of the atrophic clavicular nonunion by double-plate fixation with autogenous cancellous bone graft: a prospective study

Background The objective of this study is to assess prospectively the effectiveness of double-plate fixation combined with autogenous cancellous bone graft in the treatment for the atrophic clavicular nonunion. Methods Between February 2013 and November 2017, a total of 12 patients with atrophic clavicular nonunion (mean age, 40.4 ± 9.0 years, range, 27–60 years) were treated by double-plate fixation with autogenous cancellous bone graft in our institute. The Constant Score System was used to evaluate the preoperative and final outcomes. The Short Form-36 (SF-36) outcome questionnaire was used to assess the final clinical results. Results All patients were followed-up, with the average follow-up of 34.7 ± 6.7 months (range, 24–48 months). The healing rate was 100% in our study. The mean time of bony union was 9 weeks (range, 8–10 weeks). One patient had a postoperative superficial infection, and the patient was cured by oral antibiotics and wound dressing. No implant-related complications (plate-screw loosening or breakage) were observed postoperatively. No vascular injury, neural impairment, or thoracic outlet syndrome was discovered preoperatively or postoperatively. There is a statistical significance between the preoperative and the postoperative constant scores (P < 0.05). All the patients were satisfied with their final clinical results by SF-36 outcome questionnaire. Average scores of the physical function and bodily pain components of the SF-36 were 94.2 ± 7.3 and 92.5 ± 5.8, respectively. Conclusion Our results presented that double-plate fixation with autogenous cancellous bone graft is an effective treatment for atrophic clavicular nonunion, especially for those with a significant bone defect.


Background
Fractures of the clavicle account for 2.6-10% of all fractures [1]. Most clavicle fractures could be cured by conservative treatment, except long-term immobilizationcaused various complications, such as pain, deformity, and functional disorder of the shoulder [2]. Therefore, most patients choose surgical treatment to achieve quicker recovery of function and better alignment, especially for the patients with a comminuted clavicle fracture [2,3]. However, surgery-related complications still exist, and clavicle nonunion is one of the main complications. Olsen BS reported that the rate of clavicle nonunion after conservative treatment was between 0.1% and 5% [4], and 2.6% after surgical treatment [5].
Many surgical techniques have been described for the treatment of clavicle nonunion, including intramedullary fixation with long pins and screws [6], vascularized fibula [7], Ilizarov External Fixator [8], lag screws fixation [9], and bone grafting [4,10]; however, there still a different consensus on the best treatment, and every solution has its own set of disadvantages. For instance, the Ilizarov technique has been successfully applied in the treatment for clavicle nonunion, but it always causes pin tract infections and subjective discomfort [8]. Although intramedullary fixation with long pins has partly improved the stability of the nonunion site, many complications still exist, such as migration of the pins and pin breakage. In addition, implantation of the intramedullary pins is difficult because of the irregular anatomic morphology of the clavicle [11][12][13][14]. Therefore, more effort is needed to obtain an optimal operative treatment for clavicle nonunion.
The cardinal surgical treatment for a clavicle fracture is one-plate fixation [15]. However, the stability of clavicle nonunion treated with a single plate is unstable in some special cases, especially atrophic nonunion with a bone defect. Therefore, we attempted to use a doubleplate construct (an AO reconstruction plate or a LC-DCP plate with auxiliary minor reconstructive plate) to provide a rigid mechanical environment for the treatment of the atrophic clavicular nonunion; in addition, an autogenous cancellous bone was also used for improving the biological environment at the site of nonunion. In this study, we assessed prospectively the effectiveness of the treatment of the atrophic clavicular nonunion by double plate fixation with autogenous cancellous bone graft and aim to provide an alternative surgical method for the disease.

Methods
We analyzed all patients with clavicle nonunion after operative treatment from February 2013 to November 2017 retrospectively. Our inclusion criteria were adult atrophic clavicular nonunion, and the exclusion criteria were (1) infectious nonunion, (2) pathological nonunion, and (3) nonunion after expectant treatment. According to the inclusion and exclusion criteria, 12 cases were enrolled. Series of laboratory examinations were performed to exclude infection, including C-reactive protein level, white blood cell count, and erythrocyte sedimentation rate. The cutaneous temperature was also measured. A pathological examination of the removed nonunion bone tissue during the revision was carried out to confirm the aseptic bone nonunion. The bone nonunion was defined according to the American Food and Drug Administration (FDA) standard. The nonunion is defined as a fracture that did not heal for 9 months and had no tendency to heal for 3 months [16,17]. Before the operation, atrophic sclerosis of the fracture end and sealing of the medullary cavity were identified in X-section. In some patients, we determine nonunion by CT scan (Fig. 1).
In this study, 12 patients with atrophic nonunion were included in the study. Five patients had a right/dominant clavicular nonunion, whereas the remaining seven patients had a left clavicular nonunion. The mean age is 40.4 ± 9.04 years (range, 27-60 years). More details were listed in Table 1.
The patients were placed supine in a beach chair position. The anesthesiologist gave the patient brachial plexus and cervical plexus anesthesia. The longitudinal Fig.1 a Clavicle nonunion with plate fixation at 9 months. b Nonunion confirmed by CT scan. c Nonunion was showed in the 3D reconstruction of the clavicle incision was made along the inferior aspect of the clavicle overlying the nonunion site. After arriving at the nonunion site, sclerotic bones were debrided until the occurrence of the paprika sign (paprika sign is bleeding from the broken ends of the nonunion). It is used to determine the blood flow at the fracture end and assess the viability of the bone fragment) [18], and then, the Kwire was used to drill the medullary cavity. The cardinal plate above the clavicle was usually 9 hole AO locking reconstruction plate or LC-DCP plate (3.5 mm, 6 screws, double cortex fixation). Plates are contoured to the patient's clavicular anatomy. We used 360°sufficiently surrounding bone defect site autogenous bone graft technique. The bone autograft material was acquired from the anterior iliac crest. The auxiliary plate in the anterior-posterior plane is usually a 7-9 hole general reconstruction plate or AO LCP(2.7 mm, 4 screws, single cortex fixation). After satisfactory X-ray examination, a large amount of normal saline was used to rinse, a drainage tube was placed, then fascia, subcutaneous tissues, and superficial skin were sutured layer by layer.
Two weeks after surgery, patients were allowed to use the arm out of the sling with the assistance of a physiotherapist. Three weeks later, the patients can resume gentle exercises. Most of the patients were able to elevate the arm for above-shoulder activities at 3 weeks after the operation. Four weeks after surgery, the patients can return to weightlifting and full exercises. The final clinical results were evaluated by the constant score system and Short Form-36 (SF-36) outcome questionnaire [19,20]. The constant score could provide a so-called numerical dual assessment: the objective evaluation index including shoulder mobility and muscle strength (65 points) and the subjective evaluation index including pain and functional activity (35 points). It includes the presence of pain (15 points), functional activities (20 points), shoulder mobility (40 points), and muscle strength (25 points) [21]. A score of 100 indicates excellent shoulder function.
The data were analyzed by SPSS 19 software (IBM, Armonk, NY, USA) and the associated IRR package. In this study, the statistical analysis was carried out including the arithmetical mean, standard deviation, and 95% confidence interval of demographic characteristics and clinical outcomes of the patients. The statistical significance was analyzed for the difference between mean pre-and postoperative constant scores using Student's t test for small, dependent samples.
A value of P < 0.05 was considered to indicate statistical significance.
The follow-up for all patients is complete. The average time of follow-up is 34.7 ± 6.7 months (range, 24-48 months). The mean time of bone union was 9 weeks (range, 8-10 weeks) (Fig. 2). There was a statistical significance between the preoperative and the postoperative constant scores (P < 0.05). The preoperative and postoperative constant scores were shown in Table 1. Average scores noted on the physical function and bodily pain components of the SF-36 were 94.2 (range, 92-99) and 92.5 (range, 90-99), respectively (Fig. 3). The average scores of the constant score and SF-36 outcomes questionnaire on the patients were listed in Table 2.
The healing rate is 100% in our study. One patient had a postoperative superficial infection, and the patient was

Discussion
This is a prospective study about the treatment of the atrophic clavicular nonunion by double-plate fixation with autogenous cancellous bone graft. The present study shows that the operation acquired satisfying outcomes. The healing rate is 100% in our study. Only one patient had a postoperative superficial infection, which was cured by oral antibiotics and wound dressing.
No distinct complications emerged postoperatively. In our study, the success rate (100% union rate) and mean time of union (9 weeks) are comparable with the previous results published by Schnetzke et al. [22], Hui-Kuang Huang et al. [23], Slavko et al. [8], Kabak et al. [24], Jones et al. [25], and Sadiq et al. [26] (Table 3). Sadiq et al. reported that the incidence of clavicular nonunion was 3%. Five patients were treated with double plate, and the final healing rate was 100% [26]. The  double-plate fixation technique could achieve a higher healing rate. DPF is more effective to accelerate fracture healing, by providing a more stable mechanical environment in fracture healing. Of course, major fixation stability was realized by locking reconstruction or LC-DCP plate above the clavicle. Firstly, the superior plate could provide the main strength of the structure of double plates. Secondly, the anterior-posterior plate could immobilize bone defects and neutralize the distribution of the part of stress. Moreover, it also could avoid minor displacement of the anterior-posterior plate and strengthen the antirotation stability. The double-plate technique is advantageous in circumferential or atrophic bone loss due to the axial and torsional stable configuration. Furthermore, the bone grafting technique was also a decisive procedure in achieving a satisfactory outcome. We used the 360°autogenous bone graft technique at the site of bone loss. The bone autograft material was acquired from the anterior iliac crest. It not only can supply biomechanical connections at the defect site, but also rebuild the clavicle blood circulation [22]. Similarly, Haj et al. reported that the treatment of nonunion after limb fracture with or without double plate autograft has achieved satisfactory results [27]. In our experience, double-plate combined with autogenous cancellous bone graft technique is necessary for patients with the atrophic clavicular nonunion, especially for bone defects. An auxiliary small plate in front of the bone defect could increase the fixation stability and maintain the graft position at the nonunion site. Double plates could also allow shoulder girdle early rehabilitation exercise postoperatively due to the absolute stability. In patients with atrophic nonunion, large bone defects often existed after removal of the sclerotic bone, so it is necessary to use an autogenous bone graft to restore clavicle length and double plates to achieve rigid fixation. In addition, there are some other key technology in the surgical procedure: (1) the ends of sclerotic bone should be debrided until the observance of the paprika sign, (2) the medullary cavity should be drilled through the K-wire, and (3) the cardinal plate above the clavicle is usually 9 hole AO locking reconstruction plate or LC-DCP plate (3.5 mm, 6 screws, double cortex fixation). (4) Plates should be pre-contoured in terms of the patient's clavicular anatomy. (5) It is essential to handle 360°autogenous bone graft at a bone defect site. The bone autograft material was acquired from the anterior iliac crest. (6) The auxiliary plate in the anterior-posterior plane is usually 7-9 hole general reconstruction plate or AO LCP(2.7 mm, 4 screws, single cortex fixation). It could improve fixation stability in the anterior-posterior plane of the clavicle.
Data on the demographic characteristics of patients, treatment results, and complication rates were included in our study. However, there are still some limitations to this study. The number of patients was relatively small.
Plate revision without bone graft (2) Plate revision with iliac crest bone graft Therefore, more patients need to be included in this study to verify the effectiveness of this procedure.

Conclusion
Our study showed that double-plate with autogenous bone graft may be a prosperous and ideal alternative osteosynthesis technique for treating the atrophic clavicular nonunion, especially those with significant bone gaps. The rigid stability of the double-fixation construct allows the patient to proceed with functional rehabilitation exercise earlier and finally obtain better results. Therefore, double-plate with autogenous bone graft may be a successful salvage procedure when traditional means of achieving union were ineffective.