Development of Symptomatic Adjacent and Remote Level Compression Fractures Following Balloon-assisted Kyphoplasty in a Series of 1,318 Patients.

BACKGROUND

Balloon-assisted kyphoplasty (BAK) is a minimally invasive procedure to treat vertebral compression fractures (VCF). BAK not only restores vertebral height and corrects kyphotic deformity by cement augmentation, but it also may alter spinal biomechanics, leading to subsequent adjacent level VCFs.

OBJECTIVES

This study aims to investigate the timing, location, and incidence of new VCFs following BAK and identify the risk factors associated with their occurrence.

STUDY DESIGN

Single-institution observational study.

METHODS

A prospectively collected cohort of 1,318 patients who underwent BAK by a single-surgeon from 2001 through 2022 was analyzed. The patients had pain that was unresponsive to nonsurgical management and a VCF secondary to osteoporosis, trauma, or neoplasm. The time between the index and subsequent fracture, fracture level, number of initial fractures, age, body mass index (BMI), tobacco use, and chronic corticosteroid use were recorded.

RESULTS

Of 1,318 patients, 204 (15.5%) patients underwent a second BAK procedure an average of 373 days following BAK (range: 2-3,235 days). Third, fourth, and fifth procedures were less common (45, 12, and 6 patients, respectively). A total of 142 patients (69.6%) developed a subsequent fracture adjacent to the index level; adjacent and remote level fractures developed at different times (mean: 282 vs 581 days, P = 0.001). Patients treated for multiple VCFs in a single surgery were more likely to develop subsequent VCFs (P = 0.024) and at adjacent levels (P = 0.007). Subsequent VCFs were associated with older age (P < 0.001), women (P = 0.045), osteoporosis (P < 0.001), and chronic corticosteroid use (P < 0.001). A subgroup analysis of 812 (61.6%) patients who underwent BAK for degenerative indications revealed that osteoporosis (b = 0.09; 95% CI, 0.03-0.16; P = 0.005) and chronic corticosteroid use (b = 0.06; 95% CI, 0-0.11; P = 0.055) were associated with adjacent level fracture. For the entire cohort, almost every patient treated for both a thoracic and lumbar fracture (92.3%) developed an adjacent level second fracture (P = 0.005).

LIMITATIONS

The true incidence of post-BAK fractures may be underestimated as surveillance is not routine in asymptomatic or osteoporotic patients.

CONCLUSIONS

Symptomatic post-BAK VCFs are infrequent and may occur long after the initial procedure. Nearly two-thirds of subsequent fractures in our study occurred adjacent to the initially treated level; almost every patient who suffered thoracic and lumbar fractures at the same time developed an adjacent level second fracture. Additionally, osteoporosis and chronic corticosteroid use were associated with adjacent level fractures in patients who underwent surgery for degenerative indications.