As the population ages, the frequency of applying pedicle screw fixation is increasing in osteoporotic spine treatments. Pedicle screw loosening, migration, and pull out are the most common postoperative complications for osteoporotic patients who have undergone spine internal fixation (1, 2, 4). The incidence of fixation failure and associated complications has gradually increased, and the question of how to improve pedicle screw stability in osteoporotic spines has become a hot issue (2–4). The use of pedicle screws with cement augmentation, frequently PMMA, has recently become a reliable and feasible method for preventing pull-out due to its reliable biomechanical stability (2).
In the present study, good clinical outcomes were observed in the patients in the two groups postoperatively. However, the VAS, ODI and JOA scores of the CAPS group at three and six months after surgery improved significantly than those of the CPS group. There were four pedicle screws loosening in the CAPS group and 17 pedicle screws loosening in the CPS group. The loosening rate of the CAPS group was 0.85% (4/470), which was significantly lower than the 6.25% (17/272) rate of the CPS group.
The firm and stable fixation with cement augmentation provides the basis for early rehabilitation, so the CAPS group enjoyed a faster recovery rate and better postoperative outcomes than the CPS group in the early postoperative period. Additionally, the fusion rate of the CAPS group was better than that of the CPS group. The incidence of cage subsidence has been reported to be as high as 59% in posterior lumbar interbody fusion (20). The present study showed 40 segments with cage subsidence (38.8%) in the CAPS group and 28 (46.7%) in the CPS group, but the difference between the two groups was not significant. However, patients’ BMD was found to be significantly correlated with cage subsidence in both groups (P < 0.05). Similar to previous study (12), we also found that patients in the CAPS group achieved a significant improvement in the Taillard index compared to those in the CPS group (P < 0.05) in the patients with lumbar spondylolisthesis, and a significant difference in the intervertebral disc space height between the two groups was found at 12 months after surgery and at the final follow-up.
Cement leakage is the most frequent complication of the CAPS method (2, 21–23). The reported incidence of cement leakage in CAPS patients is highly variable and ranges from 0–80% (2). In our series, cement leakage was observed in 26 patients (26/99, 26.3%) and 34 screws (34/470, 7.23%), and all the cases were asymptomatic. The literature reports that the rate of cement leakage is not related to the volume of PMMA, but the cement leakage rate is higher in patients with lower BMD (24). The present study also found similar results in this respect, but the difference did not reach statistical significance. However, we found that patients with a diagnosis of vertebral fracture are more prone to cement leakage, and type S leakage (i.e., leaking via the segmental vein) was the most common (88.2%, 30/34).
Several studies have reported that cement augmentation may affect the load of the vertebral body, which may result in a new adjacent vertebral fracture, especially in osteoporotic patients (25, 26). To date, few clinical studies have examined the potential risk of adjacent fractures in patients who have received CAPS treatments for osteoporotic spinal diseases. In the present study, seven and two adjacent fractures were observed during the follow-up period in the CAPS group and the CPS group, respectively. The difference between the two groups was not statistically significant (P = 0.133); however, there was a tendency for a higher risk of adjacent fracture in the CAPS group than the CPS group (7/99 vs. 2/56). Further studies with large sample sizes and longer follow-up periods are needed to assess this potential complication.
Our results indicate that the CAPS technique can significantly decrease the screw loosening rate in the osteoporotic spine with good clinical outcomes, and we are of the view that it has good prospects in clinical application. However, no clear consensus has yet been reached regarding the use of CAPS, and a BMD ≤–2.5 SD is an indication in most studies. In some cases, we also found a weak attachment of the screw into the vertebral bone even though the patients had a preoperative BMD > − 2.5 during the operation. In such circumstances, we also recommend using the CAPS method to enhance fixation strength, especially in the reduction of lumbar spondylolisthesis (24). The reasons relate to the limitations of of dual-energy X-rays for BMD measurements and the degenerative changes of the spine may affect the accuracy of such measurements. Thus, for patients at high risk of osteoporosis, a more precise instrument is warranted to examine bone density, such as quantitative CT (12, 24, 27).
The current study had several limitations. First, as a retrospective study, it has some inherent limitations. Second, more spinopelvic parameters need to be analyzed. Third, many factors are known to affect cage subsidence and adjacent segment degeneration, but we could not analyze these variables in detail.