Therapeutic efficacy of Transpedicular Intracorporeal cement augmentation with short segmental posterior instrumentation in treating osteonecrosis of the vertebral body: a retrospective case series with a minimum 5-year follow-up

Limited and contradictive studies have addressed ONV management, thus our goal was to evaluate the outcomes of TCA with SSPI for treating ONV during a relatively long follow-up period (> 5 years).

This study was approved by the Ethics Committee of Honghui Hospital (No. XAHHLLSP1022), and written consent was obtained from all of the patients. A total of 46 patients with ONV who were treated by the method of TCA with SSPI between July 2010 and July 2012 were enrolled in the present study, i.e., 22 males and 24 females, with a mean age of 62.8 ± 7.11 years. Each patient had only one affected vertebral body. The vertebral fractures were as follows: T11, 3 patients (6.5%); T12, 10 patients (21.7%); L1, 12 patients (26.1%); L2, 14 patients (30.4%); L3, 1 patient (2.2%); L4, 4 patients (8.7%); and L5, 2 patients (4.3%). All these patients had a vacuum cleft of the vertebral body that was verified by preoperative plain radiographs and computed tomography (CT) scanning. The accumulation of air or liquid was verified via MRI. Two radiologists assessed the imaging examinations (the plain radiographs, CT, and MRI), and only when both agreed with the diagnosis of ONV was included in the study.

Inclusion criteria were the following: (1) patients were assessed by physical evaluations, plain radiographs, CT, and MRI; (2) definitive determinations of an intravertebral cleft, vacuum phenomenon, and gas or liquid area in fractured vertebrae showed by radiographs, CT scans or MRI; (3) patients with symptoms or signs consistent with the ONV lesion, including tenderness in the lesion area and back pain during flexion and extension; and (4) older patients with a T score under − 3.0. The indication for treatment was the absence of pain relief during the first 4 weeks following conventional treatment. The fractured vertebra with intravertebral fluid or air pockets was chosen and the area was paired with the region of pain to establish which spinal segment required treatment.

Exclusion criteria were the following: (1) history of stroke, diabetes, rheumatoid disease, dementia, significant medical condition necessitating rigorous treatment, or simple osteoporotic compression fractures without ONV; (2) multiple vertebral fractures; (3) prior vertebral fractures in the same or a nearby vertebra above or beneath the fractured vertebra and prior spinal operation and vertebroplasty; and (4) patients with neurological compromise.

Operations were conducted under general anesthesia. Patients were placed in the prone position with sponge mats bracing the upper chest and pelvis to allow maximum extension of the spinal column. A typical posterior midline strategy was utilized to reveal the posterior portion of the spine for surgical tools, including one level over and under the buckled vertebra. Hollow cement pedicle screws were implanted into the bilateral pedicles of the proximal and distal nearby vertebrae. The injection volume of polymethylmethacrylate (PMMA) cement was 1–1.5 ml per screw. For the diseased vertebra, two tunnels were placed under fluoroscopy, with one in the cleft and the other beneath the cleft (Fig. 1). Cement was injected slowly into the diseased vertebra under continuous fluoroscopy in the lateral plane. Cement injection was ceased when the cement completely filled the cleft, and the region underneath the cleft was filled with cement over 50% of the transverse diameter of the sabotaged vertebral body. The cement volume was approximately 4.5–8 ml. After the cement injection was completed, ordinary pedicle screws (35 mm in length and a diameter of 6.5 mm) were put into place. The rods were then put into place and secured after a reduction of the cement temperature. Reduction was not performed, and lamina and facet joint fusion was not implemented. After attaching the rods to the screws, the surgical opening was closed in layers. A drain was routinely inserted into the deep wound space. After the drain was removed, the patient was allowed to ambulate. Antiosteoporotic treatment was a regular option for the patients. All the patients underwent postoperative physical therapy for 6 months. Three months of total contact brace application was recommended.

Demographic, clinical, and radiologic data were gathered from medical records. Following surgery, an evaluation was conducted at established intervals: right after the operation, at 3 months, at 6 months, at 1 year, and every year thereafter. The impact on the patient’s daily life was assessed using the Oswestry Disability Index (ODI). The Visual Analogue Scale (VAS) score was used to assess the degree of back pain. Anteroposterior and lateral radiography were conducted at every follow-up visit. Transformation of the vertebral body was analyzed through serial follow-up plain radiographs. Vertebral height was quantified as height in millimeters on the vertebral borders at the anterior and posterior margins of the sabotaged vertebral body. The average vertebral body height was determined as the mean of the anterior and posterior body height. The kyphotic angle was established with Cobb’s method on a standard lateral radiograph. Each of the heights and angles was measured with the picture archiving and communication system (PACS) and the relevant computer software (PiViewSTAR 5.0, INFINITT, Seoul, Korea).

Six patients were lost to follow-up, one died due to myocardial infarction, and three had a vertebral body fracture, which left a total of 36 patients. Each of the patients had a follow-up period of at least 5 years, with a mean follow-up period of 67 ± 4.2 months. The average hospital stay was 10.4 ± 2.3 days, the average surgical time was 132 ± 15.5 min, and the average blood loss was 410 ± 30.5 ml.

Of the 36 patients, seven patients experienced complications, namely, pneumonia (two patients, 5.56%), screw loosening (two patients, 5.56%), moderate hematoma in the subcutaneous tissue (one patient, 2.78%), and cement leakage (two patients, 5.56%). No revisions were required in the patient with screw loosening because no progression was found during the long-term follow-up. Cement leakage was located at the paravertebral portion. No neurologic deficits were found to be associated with cement leakage. No revision was needed to deal with the leakage.

The preoperative VAS score was 7.4 ± 1.6. At the 3-month postoperative appointment (3.6 ± 1.3), a significant reduction was found (P = 0.01). Succeeding VAS scores throughout the postoperative follow-up were 4.8 ± 1.4 at 12 months, 5.0 ± 1.3 at 2 years, and 5.1 ± 1.2 and 5.6 ± 1.1 at 3 and 5 years, respectively. The mean VAS scores at the 2-, 3-, and 5-year follow-up visits were significantly elevated compared to the mean VAS score at 3 months. When compared with the preoperative VAS score, no significant differences were found (Table 1). The mean ODI score was 72.4 ± 4.6 preoperatively, 32.4 ± 12.6 at 3 months postoperatively, 42.4 ± 6.6 and 48.4 ± 6.6 at 1 and 2 years, and 62.4 ± 8.6 at the last follow-up appointment. While the mean ODI score was significantly reduced until 1 year following surgery, this reduction was almost lost 5 years postoperatively. No statistically significant difference was found between the 2-, 3-, 4-, and 5–year follow-up ODI scores and the preoperative score (Table 1).

Kyphosis angle had positive correlation with VAS score (Spearman r = 0.463, P <  0.0001) and ODI score (Spearman r = 0.490, P < 0.0001), but vertebral height was not found to be correlated with VAS score or ODI score (Both P > 0.05). The mean preoperative kyphotic angle was 34.3° ± 5.4. It was significantly reduced at the immediate postoperative follow-up visit (P = 0.001). There was also a significant difference between the preoperative mean vertebral height and 3 days postoperatively (P = 0.01). At 1 year postoperatively, 26 patients developed progressive vertebral height loss. The mean vertebral height was 15.2 ± 3.5 mm at the final follow-up visit. When compared with the preoperative height, no statistically significant difference was found (P = 0.73). The kyphotic angle displayed a similar change in accordance with vertebral height (Table 2). We observed caudal vertebral body height loss in 23 patients (Fig. 2).