Between April 2009 and June 2014, 282 patients with 399 vertebral lesions (thoracic vertebrae, 223; lumbar vertebrae, 176) received PKP at the First Affiliated Hospital of China Medical University (Shenyang, China) in our hospital. The majority of the patients received systemic chemotherapy or radiotherapy prior to PKP; however, the severe pain these patients experienced remained following the treatments. All the patients met the following inclusion criteria: Presence of a primary malignant tumor, which was diagnosed by pathology or cytology; grade D-E spinal cord function, according to the Frankel grading system (11); adequate hematological, hepatic, renal, neuronal and cardiac functions, and the ability to maintain a prone position for ≥2 h; a Karnofsky performance score (KPS) of ≥60; an expected survival time of ≥3 months; vertebral destruction, which was dominated by osteolytic lesions; thoracic and lumbar vertebral fractures confirmed by physical and imaging examinations; and a lesion region located below the 6th thoracic vertebra. The characteristics and tumor data of the patients are presented in Table I. All patients were informed of the future publication of this study and provided their consent. The study was approved by the Institutional Board at the First Affiliated Hospital of China Medical University.

The efficacy of PKP was evaluated using the visual analog scale for pain (VAS), KPS and QOL scores [short form with 36 questions (SF-36)]. In addition, radiographical data, including the degree of restoration of the kyphotic angle and anterior vertebral height, and the leakage of bone cement, were measured. The safety of PKP was assessed by evaluating the complications and side effects reported during or post-surgery. A clinical assessment of the patients was performed on the day prior to the surgery and at 24 h, 3 months, 6 months and 1-year post-surgery, as well as at the last follow-up date. The KPS and QOL scores were calculated 3 months after PKP, and baseline KPS and QOL scores were assessed prior to the procedure. A clinical follow-up examination of the patients was conducted by Dr Feng Chen, Dr Wei Shan and Dr Yang Gao (Department of Interventional Radiology, First Hospital of China Medical University), and diagnostic images were independently evaluated by Mr. Xiang Wang (Department of Geriatrics, The First Affiliated Hospital of China Medical University). All the patients received conventional chemotherapy following the surgery, which was specific to the primary tumor.

The following instruments were used in the PKP procedure: Puncture needles (3.2–4.0 mm in diameter), hollow diamond and solid diamond tipped needles, precision injector pressure device (Longguan Co., Jinan, China) to assess cement pressure during injection, surgical balloon, bone cement injection gun (Longguan Co.); bone cement (Heraeus Medical GmbH, Wehrheim, Germany); and surgical hammer (Beijing Operation Apparatus Factory, Beijing, China).

X-ray, computed tomography (CT) and magnetic resonance imaging (MRI) were performed prior to surgery to determine the location and number of lesions, and the degree of collapse of the vertebrae (Fig. 1). In addition, CT and MRI established if the back wall of the vertebral body was intact and if the spinal cord was compressed. PKP was guided by a Digital Subtraction Angiology (DSA) machine (GE Healthcare Bio-Sciences, Pittsburgh, PA, USA) in the interventional radiology suite following an analysis of the radiological images. Local anesthesia (2% lidocaine; 5 ml) was administered to the patient prior to the surgery. The patient was placed in the prone position and injection needles were used to puncture unilaterally or bilaterally through the pedicle of the vertebral arch (if the metastasis was located in a lumbar vertebra) or rib vertebral joints (if the metastasis was located in a thoracic vertebra).

When the needle tip reached the back wall of the vertebra (Fig. 2), 3-dimensional CT scanning was used, in combination with DSA, to ensure that the spinal cord was not punctured and the procedure was effective (Fig. 3). A surgical hammer was used to aid the needle placement into the anterior quadrant of the targeted vertebral body. Hollow and solid diamond tipped needles were used to polish the tract. Subsequently, a balloon, which was used to restore vertebral body height and create a cavity in the vertebra for the injection of bone cement, was inserted into the vertebra using fluoroscopic monitoring. The balloon was inflated and viewed using a contrast agent. If it was challenging to inflate the balloon or a portion of the balloon reached the vertebral margin, inflation was terminated and the balloon was removed (Fig. 4). Bone cement, a mixture of polymethyl methacrylate (PMMA) and non-ionic contrast medium, was injected into the vertebra using a bone cement injecting gun and fluoroscopic monitoring. When the PMMA had solidified the injection needles were removed. The volume of bone cement injected into the vertebra was 1–5 ml, with a mean of 1.9 ml for the thoracic vertebrae and 1.97 ml for the lumbar vertebrae (Figs. 5 and 6). Following the surgery, the patients were required to rest for 6 h, and during this period the vital signs, neurology, urine volume, bowel movement, sensory motor functions of the lower limbs, and alterations in routine blood tests and CD4⁺ and CD8⁺ T cell subpopulations were monitored. The patients received mannitol (250 ml injection, daily) and solu-medrol (80 mg injection, daily) for three days to treat spinal cord edema.

Continuous data were expressed as the mean ± standard deviation (SD). A comparison of continuous variables pre- and post-surgery was performed using a one-way analysis of variance. P≤0.05 was considered to indicate a statistically significant difference.

The PKP procedures performed on 282 patients with spinal metastases were successful and without severe complications. The success rate of the puncture surgery was 100%. The clinical assessment of the patients was performed at 24 h, 3 months, 6 months and 1 year post-surgery, and the last follow-up assessment was performed via outpatient review or telephone. All 282 patients underwent follow-up assessments 24 h and 3 months post-surgery, and 272 patients were evaluated at 6 months post-surgery. A total of 60 patients were lost to follow-up at 1-year post-surgery. The follow-up duration ranged between 105 days and 15 months (mean, 401 days).

For pain evaluation, the patients were asked to evaluate their degree of pain, according to VAS, which uses a score of 0–10 to indicate levels of pain between mild and severe (0, no pain; 10, severe pain). All the patients exhibited pain relief following the surgery. The pre-operative VAS was 6.85±0.97, which decreased to 3.11±1.38 24 h post-operatively (P<0.001). The VAS of the patients remained largely unaltered at 3 months (3.17±1.521; P<0.001; 282 patients), 6 months (3.58±1.63; P<0.001; 272 patients) and 1 year (4.61±1.65; P<0.001; 222 patients), and at the last follow-up (4.91±1.99; P<0.001; 219 patients) (Fig. 7A). In addition, the analgesic intake of the patients decreased following PKP (data not shown).

The physical condition of the patients was assessed by KPS. KPS score is used to assess how the disease of a patient is progressing, how it affects the daily living abilities of the patient, and determines the appropriate treatment and prognosis of the patient. A patient with a KPS score of 100, exhibits no signs and symptoms, while a score of 0 indicates patient mortality. Thus, the higher the score, the better the health status. In the present study, the pre-operative KPS was 70.42±10.83, which increased at 3 months post-surgery (80.83±10.59; P<0.001) (Fig. 7B).

QOL (SF-36) was developed by a USA Boston health study (12,13), and was based on a previous study concerning the medical outcomes (14). The QOL health questionnaire is widely used in the evaluation of clinical trials and health policies. The questionnaire consists of 9 topics: Physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, mental health and reported health transition. A patient with a score of 100, exhibits no signs and symptoms, while a score of 0 indicates patient mortality. Thus, the higher the score, the better the health status. In the present results, the pre-operative QOL score was 67.74±9.33, which then increased to 97.41±9.55 at 3 months post-surgery (P<0.001; Fig. 7C).

The kyphotic angle is defined as the Cobb angle measured between the superior endplate of the vertebra one level above the treated vertebra and the inferior endplate of the vertebral body one level below the treated vertebra, using a lateral X-ray image (Fig. 8) (15). In the present study, the pre-operative kyphotic angle was 15.68±1.46°, which decreased at 3 months post-surgery (7.71±1.51°) (P<0.05). In addition, the pre-operative anterior vertebral height (18.40±5.59 mm) increased to 24.20±6.261 mm at 3 months post-surgery (P<0.05).

Complications reported consisted of minor paravertebral leakage of bone cement, which occurred in 10 patients due to a cortical defect. Spinal cord compression, pulmonary embolism, leukopenia, decreased immunological function and nerve root compression were not detected in any patient.