Comparison of the Vertebral Kyphotic Angle and Vertebral Body Height Ratio With and Without Adjacent Vertebral Fractures in the Early Postoperative Period After Balloon Kyphoplasty

Article information

J Minim Invasive Spine Surg Tech. 2025;10(Suppl 2):S288-S295

Publication date (electronic) : 2025 July 31

doi : https://doi.org/10.21182/jmisst.2024.01956

Hiroshi Kawaguchi^,1, Koichiro Ono¹, Daisuke Fukuhara¹, Takao Nakajima^1,2, Tokifumi Majima¹

¹Department of Orthopaedic Surgery, Nippon Medical School, Tokyo, Japan

²Department of Orthopaedic Surgery, Nippon Medical School Chiba Hokusoh Hospital, Chiba, Japan

Corresponding Author: Hiroshi Kawaguchi Department of Orthopaedic Surgery, Nippon Medical School, Sendagi 1-1-5, Bunkyo-ku, Tokyo Japan Email: hiro-kawaguchi@nms.ac.jp

Received 2024 December 8; Revised 2025 February 23; Accepted 2025 April 7.

Abstract

INTRODUCTION

Osteoporotic vertebral fractures (OVFs) are prevalent among older adults, significantly impairing activities of daily living due to lower back pain. These fractures also increase the risk of systemic complications and mortality [1,2]. While conservative management is the first-line treatment for OVF, resulting in favorable outcomes in many cases, some patients suffer from persistent lower back pain due to spinal kyphosis and instability, often caused by nonunion or severe vertebral body collapse. In such cases, surgical intervention becomes necessary. However, the decision to select invasive procedures is often complicated by the presence of multiple comorbidities in older patients.

Balloon kyphoplasty (BKP), a minimally invasive surgical technique, is widely employed for patients who do not respond to conservative treatment and offers rapid pain relief in the early postoperative period [3-5]. Despite its benefits, the vertebrae augmented with polymethylmethacrylate become stiffer than the adjacent nonaugmented vertebrae. Both theoretical and experimental studies suggest that this increased stiffness can shift biomechanical loads, further enhancing the risk of fractures in adjacent vertebrae. Consequently, adjacent vertebral fractures (AVFs) are a common complication following BKP, necessitating careful consideration. The recurrence of lower back pain associated with AVF often leads to further kyphosis and spinal instability [6].

Age, bone mineral density (BMD), and history of OVF are risk factors for the development of AVF. However, the precise mechanisms underlying AVF following BKP remain inadequately understood [7-11]. Elucidating the pathogenesis of AVF is crucial for improving surgical outcomes in BKP. Given the frequent occurrence of AVF in the early postoperative period, we aimed to investigate radiographic changes in fractured vertebral bodies from the preoperative to early postoperative stages to identify risk factors of AVF following BKP.

MATERIALS AND METHODS

This study was conducted in compliance with the principles of the Declaration of Helsinki. This study’s protocol was approved by the Institutional Review Board (IRB) of Nippon Medical School (IRB No. B-2023-820). Informed consent was not required, given the retrospective analysis of previously collected de-identified information. We included patients with osteoporotic vertebral compression fractures, while those with vertebral fractures caused by high-energy trauma or pathological fractures due to infection or tumors were excluded. Sixty-three patients who underwent BKP within 60 days of the onset of a single-level OVF and were followed up for at least 3 months postoperatively were included in this study. Postoperatively, all patients were advised to wear either hard or soft back braces out of bed to ensure spinal stability. Being out of bed was permitted the day after the surgery. The definition of AVF based on the time of onset after BKP was not established, and it was difficult to determine whether a fracture occurring in the adjacent vertebral body after BKP was due to BKP or not. Therefore, in this study, AVF was defined based on previous studies; that is, fractures occurring within 3 months post-BKP were considered adjacent fractures [12,13]. Specifically, AVF was diagnosed when fractures occurred in the vertebra immediately superior or inferior to the BKP vertebra within the period. AVF diagnosis was based on lateral lumbar radiography indicating vertebral instability or magnetic resonance imaging revealing vertebral bone marrow edema. Patients were classified into 2 groups: AVF and non-AVF groups.

Comparative analyses were conducted on variables including age, sex, body mass index (BMI), and BMD, measured at the lumbar spine (L1–4) and averaged out, history of OVF, the volume of bone cement injected during BKP, and lumbar spine radiographic findings captured preoperatively, immediately postoperatively, at 1- and 2-week postsurgery, and during the final follow-up. Specific radiographic parameters, including vertebral kyphotic angle (VKA), defined as the angle formed between the upper and lower endplates of the fractured vertebra, and anteroposterior vertebral height ratio (A/P ratio), calculated as the ratio of the anterior height to the posterior height of the vertebral body, were meticulously compared between the AVF and non-AVF groups (Figure 1).

Figure 1.

Method for determining the vertebral kyphotic angle (VKA) and anteroposterior vertebral body height ratio (A/P ratio). VKA, angle between the upper and lower end plates of the fractured vertebra (angle A); A/P ratio=anterior vertebral body height (B)/posterior vertebral body height (C) × 100 (%).

All statistical analyses were performed using IBM SPSS Statistics ver. 29.0 (IBM Co., USA). The Mann-Whitney U-test was employed to evaluate continuous variables; categorical data were compared using Fisher exact test. A p-value less than 0.05 was considered statistically significant.

RESULTS

The study included 63 patients with an average age of 81.6 years; 21 patients were males, and 42 patients were females. Among them, 31 patients had a documented history of OVF. The mean follow-up duration after BKP surgery was 9.6 months. The distribution of fractured vertebral bodies was as follows: T11 (n=5), T12 (n=19), L1 (n=19), L2 (n=6), L3 (n=10), L4 (n=3), and L5 (n=1). In cases where the vertebral body was both fractured and flattened, causing collapse of both the anterior and posterior walls, neither the VKA nor the A/P ratio adequately captured the extent of vertebral deformation. Notably, no cases in this study showed the vertebral body exhibited clear flattening.

AVF occurred in 8 of the 63 patients, representing 12.7% of the cohort. The onset of AVF ranged from 2 to 12-week post-BKP surgery, with a mean onset time of 5.3 weeks. In 3 of the 8 AVF cases, BKP was repeated on the AVF-affected vertebral bodies due to recurrent low back pain.

No significant differences were observed between the AVF and non-AVF groups regarding patient demographics or the volume of bone cement used during surgery (Table 1).

Table 1.

Clinical and demographic data of patients with and without adjacent vertebral fracture (AVF)

VKA deteriorated in both groups by the final follow-up. However, the VKA worsened earlier in the non-AVF group. A statistically significant difference was observed between the 2 groups at 2 weeks postoperatively, with VKA at 17.3° in the AVF group versus 15.2° in the non-AVF group before surgery (p=0.322), 6.4° versus 6.8° immediately after surgery (p=0.489), 7.1° versus 9.6° at 1 week postoperatively (p=0.148), 6.6° versus 10.7° at 2 weeks postoperatively (p=0.047), and 8.7° versus 12.2° at the final follow-up (p=0.176) (Figure 2).

Figure 2.

Comparison of changes in the vertebral kyphotic angle with and without adjacent vertebral fracture. Values are presented as mean±standard deviation. AVF, adjacent vertebral fracture; VKA, vertebral kyphotic angle. ^*p<0.05, statistically significant differences.

A/P ratio also deteriorated in both groups over the postoperative course. However, no significant difference was observed between the 2 groups at any time point. Specifically, the values were 53.1% versus 62.9% before surgery (p=0.127), 89.0% versus 85.0% immediately after surgery (p=0.543), 81.3% versus 78.2% at 1 week postoperatively (p=0.302), 81.7% versus 75.5% at 2 weeks postoperatively (p=0.215), and 75.4% versus 72.2% at the final follow-up (p=0.496) (Figure 3).

Figure 3.

Comparison of change in A/P ratio with and without adjacent vertebral fracture. Values are presented as mean±standard deviation. AVF, adjacent vertebral fracture; A/P ratio, anteroposterior vertebral body height ratio.

The magnitude of change in VKA and A/P ratio from preoperative to immediate postoperative measurements was greater in the AVF group, with change in the A/P ratio differing significantly between the groups. Specifically, the VKA change was 10.9° in the AVF group versus 8.4° in the non-AVF group (p=0.278), and the A/P ratio change was 35.9% in the AVF group versus 22.2% in the non-AVF group (p=0.037). These findings indicate that the fractured vertebrae were more substantially corrected in the AVF group compared with the non-AVF group (Table 2).

Table 2.

Comparison of vertebral kyphotic angle and anteroposterior vertebral height ratio between the 2 groups

DISCUSSION

In this study, we identified excessive restoration of fractured vertebral body height as a significant risk factor for AVF post-BKP surgery. Early postoperative recompression of fractured vertebrae was predominantly observed in the non-AVF group. Importantly, this is the first report to confirm that recompression of fractured vertebrae can occur as early as 2 weeks postoperatively in the non-AVF group compared with the AVF group.

Excessive restoration of vertebral body height through balloon dilatation and cement injections has been well-documented as a risk factor for AVF [7,10,14]. Baroud et al. [15] reported that the injection of cement into the fractured vertebra increases the medial bulge of the adjacent vertebral endplates, potentially causing AVF after percutaneous vertebroplasty in osteoporotic patients. Similarly, BKP, which utilizes a balloon to elevate the endplate of a fractured vertebra followed by cement injection, can affect adjacent vertebral bodies through a similar mechanism. Several studies have suggested that BKP can increase mechanical stress on adjacent vertebral bodies, thereby contributing to AVF development [6,16].

The literature supports the notion that excessive correction through BKP can increase the risk of AVF due to increased mechanical stress on adjacent vertebrae and intervertebral discs [6,16]. Theoretically, increased stress results from the disproportionate restoration of vertebral body height, resulting in altered biomechanics and potential overloading of adjacent structures. Our study demonstrated that early postoperative recompression of fractured vertebrae was significantly less in the AVF group compared to the non-AVF group, suggesting that AVF may occur because the load on adjacent vertebrae is not adequately reduced by recompression of the fractured vertebral body.

Local kyphosis correction achieved by BKP tends to be lost during postoperative AVF [17]. However, in our study, there was no clear evidence that loss of correction reduced AVF occurrence. This suggests that while regional spinal alignment improved by BKP may be compromised due to AVF or recompression of the fractured vertebra in the early postoperative period, maintaining spinal alignment with BKP alone is challenging.

The instability of the fractured vertebrae caused by OVF contributes significantly to lower back pain, making the stabilization of the affected segment with cement injection crucial. Given the difficulty in maintaining local spinal alignment with BKP alone, focusing on immobilizing the instability of the fractured vertebral in OVF treatment for older patients rather than achieving a precise reduction of vertebral body height is important [18].

AVF occurs relatively frequently in the early postoperative period, with reported incidence rates ranging from 9% to 30.9% [9,10,19,20]. Several studies reported that the incidence of AVF could be reduced by performing BKP surgery early after injury, supported by significantly lower AVF rates in patients who underwent BKP within 4 weeks of injury. These studies suggested that early BKP could prevent changes in the local biomechanical environment caused by the kyphotic collapse of vertebral compression fractures [20,21]. In our study, AVF occurred in 8 out of 63 patients (12.7%) who underwent BKP within 60 days of injury, which is a relatively lower frequency than reported previously [9,10,19]. Early BKP following the onset of OVF effectively prevents AVF.

We found a higher occurrence of OVF and AVF at the thoracolumbar junction, consistent with previous reports [22]. The imbalance in fracture location may have influenced the results. A matched cohort study can provide stronger evidence to identify the causes of AVF. However, due to the small sample size in this study, such an analysis was not feasible herein. Future research with a larger cohort should incorporate a matched study design to further validate these findings.

In previous studies, various factors, including sex, age, previous OVF, and BMD, among others, have been identified as risk factors for AVF [10,11,17,20,21,23]. Matsumoto et al. [10] reported that age, previous vertebral fractures, and local kyphosis were associated with a higher incidence of AVF, particularly in patients aged 75 years or above and those with multiple existing vertebral fractures. Additionally, Gong et al. [23] found that the AVF segment computed tomography Hounsfield unit (HU) was significantly lower in AVF cases compared with those that did not subsequently develop vertebral fractures after surgery, highlighting low HU as a significant risk factor for AVF. HU measurement was not analyzed in this study and is an issue for future investigation. Similarly, Movrin et al. [11] identified preoperative BMD as a key risk factor in patients undergoing PVP and BKP for OVF, reporting a correlation between lower BMD and increased AVF incidence. Low BMI is correlated with low BMD [24]. In contrast, we did not find significant differences in patient demographics, including age, BMI, and BMD, between the AVF and non-AVF groups. This discrepancy may be attributed to the higher average age and lower BMI and BMD of our study cohort compared with those in the aforementioned reports, which could have minimized the impact of these variables on AVF occurrence in our population.

This study has some limitations that merit further consideration. First, it was a retrospective analysis, which may have introduced selection and indication biases for BKP and nonoperative management. Second, the study did not consider variations in osteoporosis treatment, as the management approach was individually determined by the attending physician and ranged from no intervention to oral therapy or injection. Antiresorptive and anabolic drugs could influence fracture risk. However, a standardized osteoporosis treatment protocol was not applied. Third, intra- and interobserver differences were not calculated for radiological measurements. Fourth, we did not include a non-BKP control group, which limits our ability to assess the absolute risk of AVF following BKP. In patients with osteoporosis and a history of OVF, the incidence of subsequent new OVF is considerable and comparable to the rate of AVF observed after BKP [25]. Furthermore, other studies indicate that the incidence of AVF is similar regardless of the BKP surgery [26]. Future studies incorporating such comparisons are warranted to further validate our findings. Despite its limitations, we found that fractured vertebral bodies in the AVF group were more substantially corrected by BKP, and correction loss was more prominent in the non-AVF group from the early postoperative period. Notably, early postoperative loss of correction occurred predominantly in the non-AVF group, suggesting that AVF may be induced because the load on adjacent vertebrae is not sufficiently reduced by the recompression of the fractured vertebral body during the early postoperative period. Excessive correction through BKP can increase the pressure on adjacent vertebrae, further heightening the risk of AVF. Conversely, moderate correction loss may reduce this pressure, potentially lowering AVF incidence. These findings underscore the significant impact of pressure changes on adjacent vertebrae induced by BKP, highlighting the need for careful consideration of vertebral body correction during the procedure to mitigate the risk of AVF.

CONCLUSION

In summary, this study demonstrated excessive correction of vertebral body height during BKP as a significant risk factor for AVF post-BKP surgery due to increased mechanical stress on neighboring vertebrae. Our findings suggest that moderate correction loss in the early postoperative period may alleviate this risk, underscoring the need for careful consideration of vertebral body correction during BKP to minimize AVF occurrence and optimize patient outcomes.

Article information Continued