This retrospective study was approved by the local ethics committee. From January 2009 to December 2016, 18 children who had an SBC of the humerus were treated with intramedullary decompression and drainage with ESIN combined with intralesional injections of steroids. The surgery-indications of these patients included large and painful SBCs with or without pathological fracture. Diagnosis information was obtained from orthopedic files, including preoperative/postoperative X-rays, computed tomography (CT) and magnetic resonance imaging (MRI). Clinical data included gender, age, symptoms, presence or absence of pathological fracture, surgical procedures, and functional or radiological outcomes.
As well known, X-ray images of SBCs show that the medullary cavity is a central elliptical bright shadow with no gravel-like densification point inside, which sometimes is segregated by bone ridges, and cortical bone will expand and becomes thinner, but there is no periosteal reaction (except pathological bones). Magnetic Resonance Imaging (MRI) always presents a low or intermediate signal on T1-weighted images and a homogeneous high signal on T2 weighting. The cystic fluids extracted by surgeon are tested for pathological examination, which can confirm the diagnosis of SBCs. In this study, considering the diagnosis of simple bone cysts was clear based on related images, especially in X-rays and MRI, no preoperative biopsy was performed. Moreover, the clinical and radiologic features were used for differential diagnosis of SBC from another cystic lesions based on a previous report , including aneurysmal bone cyst, fibrous dysplasia, enchondroma, eosinophlic granuloma and intraosseous ganglia. Briefly, aneurysmal bone cyst on roentgenograms appear as a lytic, eccentric, intramedullary bone lesion, with a transverse diameter that is wider than the epiphyseal plate, and the MRI images of these lesions show double-density fluid levels and septations. Fibrous dysplasia cases can be distinguished by ground glass appearance of the matrix. Enchondromas are distinct radiolucent intramedullary lesions with thinning and expansion of the cortices, which are usually happened in short tubular bones of the hands and feet. Eosinphilic granuloma frequently involves axial skeleton than appendicular skeleton, while intraosseous ganglia are small radiolucent lesions that mainly observed in the epiphysis and subchondral region.
The demographic data of this study group are summarized in Table 1. 11 males and 7 females were included in this study, and the mean follow-up period was 40 months (range, 19-65 months). Most of the cysts were located in the metaphyseal, isolated diaphyseal or metaphyseal-diaphyseal regions of the humerus. According to the standard proposed by Neer et al , the cysts were distinguished into four grades, and the classification was based on the severity of the lesion. As a result, those cysts were found active in 16 cases and inactive in 2 cases. Most of the patients were brought to the outpatient by their parents due to upper arm pain or accidental injury, or diagnosed pathological fracture in other institutions. A pathological fracture happened in 12 cases. Recurrence, partial healing and pathological fracture were all our surgical indications.
After a review of related imaging studies, according to symptoms and physical signs, a conclusion that a benign tumor was the more likely diagnosis was drawn. Surgery was always performed under general anesthesia and radiographic control, and it started from an incisional penetration with a big syringe in the region of the bone cyst located under a C-arm X-ray. The order of the penetration was from the distal part of the cast to the proximal and the surgeon should try to avoid touching vital nerves and vessels in case of hurting any of them. The syringe was through minimally percutaneous penetration, trying to avoid open incision. If the cystic cavity was too large, two or three penetrative points were necessary. Then extracted the cyst liquid which was yellow and transparent, and hemorrhagic combined with pathological fractures. We extracted the fluid in the cyst with a 5ml medical syringe and send it to histopathologic examination. Afterwards, wash the cavity with normal saline and cause no further damage to the wall.
The fluid in the cyst was centrifuged, smeared onto a slide, evaluated by H&E staining and observed under microscope (40 ×, 100 × and 200 ×). Patients were diagnosed based on their symptoms as well as the results of X-ray, CT, MRI, and pathology.
Titanium elastic intramedullary nailing (TEN) was applied, which meant to insert elastic intramedullary nails through windows cut on the lateral cortex of the distal of humerus. There were two operative approaches. One was to operate on the medial epicondyle of the humerus and ectepicondyle of humerus, and the other was to operate on the same side of the lateral of ectepicondyle of humerus. The surgeon should be careful not to cause any ulnar nerve injure. For the patients with pathological fractures, reduction should first be performed to reduce injury. The length of the nails was variably selected according to the patient’s sex, age, and the bone length (confirmed on the basis of the preoperative images). The diameter of the nails was selected according to the criterion which said 2 nails would occupy 2/3 of the minimum diameter of the medullary cavity, and the longest one was not allowed to be beyond the epiphyseal plate line. In case of disturbing epiphyseal growth, the distal end of the nails was left in a manner to avoid irritation of the surrounding soft issues. The procedure was under the guidance of a C-arm system. As the elastic intramedullary nail passed through the cyst, decompression and drainage were completed.
Methylprednisolone acetate was injected into the cavity through previous percutaneous penetrative point at a variable dose according to the volume of the cavity. 200 to 2000 mg of methylprednisolone acetate (40mg/ml) was injected into the humeral cavity. Since the elastic intramedullary nail had passed through the bone cyst and the decompression was done, the internal drainage was accomplished.
Postoperative patient management
All the patients wore a sling after operation, the patient was checked every 2 days. On average, it took about 7-14 days for them to stay in hospital. Active finger and waist motion, and passive elbow and shoulder motion were allowed immediately after operation. Active elbow and passive shoulder motion were allowed 4 weeks after operation. Active shoulder motion was allowed 6-8 weeks after operation.
Removal of the intramedullary nailing
For the well-healed SBC patients, the elastic intramedullary nails were removed as soon as possible. As the protocol to remove the lastic intramedullary nail in the treatment of children humerus fracture, we expanded the original incision to expose the elastic intramedullary nails while protecting the protect local soft tissue, blood vessels and nerves. Then clenched the end of the elastic intramedullary nails with pliers, knocked gently along the long axis of the humerus longitudinal direction to loss the nails, and pulled out elastic intramedullary nail in the opposite direction. For the nails that unable to observed accurately, we would place a 5 ml syringe needle nearby, and search for the nail by intraoperative fluoroscopy (C-arm), remove part of the bone cortex to expose the nail, and then remove it as described above.
Radiological and functional analysis
Radiological and functional follow-ups were mostly taken in the orthopedic outpatient clinic, and patients' radiographs were evaluated at admission (preoperative), 1 week, 1 month, 2 months, 3 months, 6 months, 12 months and 24 months after the operation, as well as the last follow up. All the patients were asked to take an anteroposterior and lateral radiograph of the humerus.
Musculoskeletal Tumor Society (MSTS) criteria was used to assess the function  before and after the operation (Table 2). Using this scoring system, each patient’s emotion, function and pain were evaluated, besides, weight lifting for upper lesions, hand position and hand skills were also recorded.
Treatment success was evaluated by Capanna criteria , which includes four grades, grade 1: complete healing, fully filled with bone; grade 2: partial healing with a small residual cystic area remains; grade3: partial healing with a large residual cystic area remains; grade 4: partial healing, with response (Table 3). In order to make statistical analysis easier, we made a slight revision of Capanna criteria. As shown in Table 6, we defined grade 1 to be 4 points, grade 2 to be 3 points, grade 3 to be 2 points, and grade 4 to be 1 point. Preoperative and postoperative results of the last follow-up were recorded in Table 7.
Early or late complications were recorded, including wound problems, infection, refracture, deformity and nerve injury.
Statistics work was done with SPSS17.0 statistical software (USA). Paired t tests were used to compare the MSTS scores, the visual analog scale (VAS) scores and the Capanna scores separately before and after operation. The statistically significant difference level was set at * p < 0.05, **p < 0.01 and ***p < 0.001.