Clinical outcome of vertebroplasty alone versus short-segment posterior instrumentation with vertebroplasty in osteoporotic vertebral fracture: a propensity-score-matched analysis

Article information

Asian Spine J. 2025;19(1):28-37

Publication date (electronic) : 2025 February 4

doi : https://doi.org/10.31616/asj.2024.0231

Borriwat Santipas, Nath Adulkasem, Korawish Mekariya, Ekkapoj Korwutthikulrangsri, Monchai Ruangchainikom, Werasak Sutipornplalangkul

Department of Orthopaedic Surgery, Siriraj Hospital, Mahidol University Faculty of Medicine, Bangkok, Thailand

Corresponding author: Borriwat Santipas, Division of Spine Surgery, Department of Orthopaedic Surgery, Siriraj Hospital, Mahidol University Faculty of Medicine, 2 Wanglang Road, Bangkok noi, Bangkok 10700, Thailand, Tel: +66-2-419-7969, Fax: +66-2-419-7961, E-mail: Borriwat.san@mahidol.ac.th

Received 2024 June 12; Revised 2024 November 11; Accepted 2024 November 13.

Abstract

Introduction

Osteoporotic vertebral fractures (OVFs) are a predominant concern, particularly among the aging population. These fractures cause intense pain, diminished mobility, increased disability risk, and reduced health-related quality of life [1,2]. The global incident increased by 38% from 6.2 million in 1990 to 8.6 million in 2019 as the global population ages [3]. Consequently, the direct and indirect expenses related to managing these fractures have exhibited a significant increase over time [4,5].

The optimal treatment approach for OVFs remains unclear. Typically, treatment begins with non-surgical interventions, encompassing initial bed rest, pain-relieving medications, physiotherapy, and orthosis utilization in specific cases. Surgical treatments provide a broad spectrum of options, including vertebroplasty (VP)/kyphoplasty, short-segment instrumentation (SS), and long-segment instrumentation. These are administered alone or in combination with anterior reconstruction and cement augmentation [6].

In 2018, the working group “osteoporotic fractures” of the Spine Section of the German Society for Orthopaedics and Trauma introduced a classification system (OF classification) for osteoporotic thoracolumbar fractures [7]. Additionally, they provided treatment recommendations according to fracture morphology and Osteoporotic Vertebral Fracture Score (OF-score) [6]. In recent times, numerous studies have investigated OVF treatment utilizing the OF-score [8–10]. Both the OF classification and score have been proven as suitable fracture severity indicators. Furthermore, they have demonstrated a significant positive correlation with surgical procedure invasiveness [11].

The classification indicated the distribution of the fracture morphologies at the thoracolumbar spine with predominant OF3 fractures (42%) followed by OF4 fracture (27%), OF2 (26%), and only minor numbers of OF1 and OF5 fractures [11]. Surgical treatment was recommended in OF3–5 with different options. Posterior instrumentation with an option of cement augmentation of the fractured vertebral body or even stand-alone cement augmentation of the fractured vertebral body in mobilized patients was recommended for the OF3. Posterior instrumentation with cement augmentation, long-segment posterior instrumentation, or posterior instrumentation with an option for anterior reconstruction for the OF4 fractures. However, the final treatment decision depends on the physician’s discretion.

Previous studies indicated the comparison of the efficacy of SS versus long-segment instrumentation [12], posterior instrumentation with and without anterior reconstruction [13], and kyphoplasty versus percutaneous posterior instrumentation [14]. To the best of our knowledge, no previous report focused on the efficacy comparison of VP alone versus SS with VP.

Materials and Methods

Results

Study population (unmatched cohort)

We determined 60 patients who were treated with VP or SS with VP at Siriraj Hospital, Bangkok, Thailand from January 2019 to December 2021. Of these patients, 19 underwent SS with VP and the rest underwent VP alone. The average age was 73.89±5.79 years in the SS+VP and 74.76±8.760 years in the VP groups (p=0.653). Most patients were female in both groups (68.4% and 87.8%, p=0.122). Of these patients, four were classified as OF-3 and 56 as OF-4. The average OF-score was 9.11±1.66 in the SS+VP and 8.15±1.93 760 in the VP groups (p=0.771) which indicated surgical treatment by OF-classification recommendation. The most predominant fracture location was T12 and L1 levels in both groups. Other baseline characteristics of patients were not different between the two groups. Table 1 shows the details of baseline characteristics.

Table 1

Baseline characteristic

Perioperative details and radiographic outcome (unmatched cohort)

The perioperative details and radiographic outcomes for the unmatched cohort reveal significant differences between the two treatment groups (Table 2). Patients undergoing SS+VP demonstrated a notably longer mean operative time of 141.32±47.75 minutes compared to 31.41±8.38 minutes for those undergoing VP alone (p<0.0001). Additionally, blood loss was significantly higher in the SS+VP group (302.63±230.60 mL versus 5.71±2.98 mL, p<0.0001). The SS+VP group exhibited higher reoperation (15.8% versus 4.9%) and complication rates (15.8% versus 7.3%), but these differences were not statistically significant (p=0.160 and 0.317, respectively). One patient in the SS+VP group underwent an extension of fusion to long instrumentation due to fracture progression just 1 month after the initial surgery. Additionally, two other patients needed debridement for superficial wound infections during the same admission. One patient in the VP group initially underwent VP at L1 and then presented with a new fracture at T10, 9 months later, for which they received another VP. Another patient first underwent VP at T12 and then, 10 months later, required VP at T9 due to a new fracture. No serious complication was reported in both groups. Three patients in the VP alone group exhibited asymptomatic cement leakage. Two patients in the SS+VP had superficial wound infections. The SS+VP group achieved greater sagittal Cobb angle correction (10.63°±6.34° versus 6.07°±6.13°, p=0.01). Progressive kyphosis at 3 months was observed in 36.8% of the SS+VP group versus 22% of the VP group (p=0.368), and at 1 year in 57.8% of the SS+VP group versus 31.7% of the VP group (p=0.100). Adjacent fractures occurred in both groups (adjacent fracture incidence rate at 1 year was 21.1% in the SS+VP group compared to 12.2% in the VP group [p=0.380]), but they were conservatively managed with minimal pain reported by the patients. These fractures were determined radiographically due to further collapse at the adjacent levels.

Table 2

Radiographic outcome and perioperative details

Quality of life (unmatched cohort)

The unmatched cohort demonstrated significant differences in the quality-of-life outcomes between the two groups at various time points, although both groups experienced improvement (Table 3). Preoperatively, the ODI was 62.34±18.12 for the SS+VP group and 57.67±20.92 for the VP group, with no significant difference (p=0.335). However, at 1 year postoperatively, the SS+VP group demonstrated a significantly lower ODI score (30.38±17.12) compared to the VP group (43.99±20.44), with a p-value of 0.012. The preoperative EuroQol-5D (EQ-5D) scores were 0.28±0.31 for the SS+VP group and 0.43±0.33 for the VP group, with no significant difference (p=0.053). The SS+VP group demonstrated a significantly higher EQ-5D score (0.80±0.19) compared to the VP group (0.66±0.26) at 1 year postoperatively (p=0.037). The mean change between preoperative and 1-year postoperative ODI scores was significantly greater in the SS+VP group (31.96±26.86) compared to the VP group (13.69±20.22, p=0.005). Similarly, the mean change in EQ-5D scores exhibited a significant improvement in the SS+VP group (−0.53±0.34) compared to the VP group (−0.22±0.33, p=0.002).

Table 3

Quality of life

Study population (matched cohort)

After matching patient cohorts across various covariates, 19 patients in both the SS+VP and VP groups were included in the subsequent analyses. Patient demographics, comorbidity burden, and operative characteristics were substantially balanced between the groups (Table 1). The mean age at surgery for the SS+VP group was 73.89±5.79 years compared to 69.74±7.54 years for the VP group (p=0.098). The proportion of female patients was 68.4% and 57.89% in the SS+VP and VP groups, respectively (p=0.737). Both groups exhibited similar body mass index values, with means of 23.96±4.01 kg/m² for the SS+VP group and 23.07±3.53 kg/m² for the VP group (p=0.4415). The proportion of patients receiving current osteoporosis treatment was 15.8% and 21.05% in the SS+VP and VP groups, respectively (p=1.000). Additionally, 68.4% of the SS+VP group demonstrated a history of trauma compared to 57.89% of the VP group (p=0.737). The baseline characteristics were well-matched, ensuring a fair outcome comparison (Fig. 1, Table 1).

Fig. 1

The density plot illustrates the distribution of propensity scores for patients undergoing vertebroplasty (VP) alone versus short-segment posterior instrumentation with vertebroplasty (SS with VP), both before and after matching.

Radiographic outcome and perioperative details (matched cohort)

The matched cohort analysis revealed significant differences in operative and postoperative outcomes between the two groups (Table 2). The mean operative time was significantly longer for the SS+VP group at 141.32±47.75 minutes compared to the VP group at 31.47±8.00 minutes (p<0.0001). Additionally, blood loss was significantly higher in the SS+VP group at a mean of 302.63±230.60 mL compared to that in the VP group at 5.58±2.06 mL (p<0.0001). However, the reoperation rate was not significantly different, with 15.79% and 5.26% in the SS+VP and VP groups, respectively (p=0.604). The complication rate was similar between the groups, with 15.79% and 10.53% in the SS+VP and VP groups, respectively (p=1.000).

Radiographic outcomes revealed that the SS+VP group demonstrated greater sagittal Cobb angle correction (mean 10.63±6.34°) compared to the VP group (mean 5.74±5.91°, p=0.0188). Progressive kyphosis at 3 months was observed in 36.84% of the SS+VP group compared to 15.79% of the VP group (p=0.269). Progressive kyphosis at 1 year was observed in 57.89% and 26.32% in the SS+VP and VP groups, respectively (p=0.099). The incidence of adjacent fractures at 1 year was 21.05% and 5.26% in the SS+VP and VP groups, respectively (p=0.34).

These results indicate that the SS+VP group experienced longer operative times and greater blood loss, but they achieved superior sagittal angle correction. However, both groups demonstrated similar rates of reoperation, complications, progressive kyphosis, and adjacent fractures, indicating that both treatment modalities are viable options with specific benefits and drawbacks (Fig. 2).

Fig. 2

(A–F) Comparison of preoperative, postoperative, and 1-year postoperative of both procedures. (B) Immediate postoperative X-ray after vertebroplasty (VP). (C) 1-year postoperative X-ray after VP. (E) Immediate postoperative X-ray after short-segment posterior instrumentation with VP (SS with VP). (F) 1-year postoperative X-ray after SS with VP.

Discussion

A narrative review by Jang et al. [15] investigates various management strategies for OVFs, highlighting the absence of a universally accepted treatment standard. Conservative methods, such as pain management, bracing, and exercise, are typically first-line treatments, but they frequently fall short for more complex cases. Vertebral augmentation techniques, including VP and kyphoplasty, provide pain relief but are debated due to related complications. Surgical interventions are required for severe cases and must be customized to address specific instability and deformity issues. Additionally, effective OVF management involves treating underlying osteoporosis with medications, such as bisphosphonates and teriparatide, though their effect on fracture healing requires further investigation.

OF types 3 and 4 fractures emerged as the most prevalent types of osteoporotic vertebral body fractures. Spiegl et al. [16] revealed that OF type 3 fractures, characterized by deformation of one endplate with significant posterior wall involvement, accounted for 65.6% of cases. Similarly, they reported OF type 3 fractures as the most predominant, comprising 42% of their large multicenter cohort, followed closely by OF type 4 fractures, which involved both endplate deformation and accounted for 27% of cases in another study [11]. OF-3 and OF-4 osteoporotic vertebral fracture treatment spans a spectrum from conservative management to advanced surgical techniques, reflecting the need for individualized approaches according to fracture severity and patient health. Conservative treatments, including pain management, bracing, and physical therapy, are suitable for less severe cases or patients with high surgical risks but may result in higher complication rates over time. Minimally invasive surgical techniques, such as VP, kyphoplasty, and SS with cement augmentation, provide effective stabilization with reduced recovery times and are predominantly utilized for OF-3 and OF-4 fractures.

Our study provides a comprehensive comparison between VP alone (VP) and SS+VP for treating OVFs, specifically OF type 4. The results indicate significant differences in radiographic and quality-of-life outcomes between the two treatment modalities. SS+VP demonstrated superior sagittal angle correction and better patient-reported outcomes, such as the ODI and EQ-5D, 1 year postoperatively. However, recognizing that posterior fixation may restrict lumbar spine movement, which could affect mobility and functional outcomes over time, is essential. The superior quality-of-life scores observed in the fixation group reflect immediate postoperative improvements, but longer-term studies are warranted to evaluate whether or not these functional benefits are sustained or offset by potential mobility limitations associated with lumbar fixation. Additionally, VP alone is frequently performed under local anesthesia, making it a suitable option for patients with higher anesthesia risks, whereas SS with VP typically requires general anesthesia due to the increased invasiveness. Furthermore, these advantages of SS+VP come at the cost of higher intraoperative risks, including increased blood loss and longer operative times. SS+VP exhibited similar rates of perioperative complications, kyphotic progression, adjacent fractures, and reoperation rates compared to VP alone despite the increased risks.

Both VP and SS+VP effectively corrected sagittal alignment immediately postoperatively. The better initial sagittal Cobb angle correction in the SS+VP group can likely be attributed to Cobb angle reduction while the patient is in the prone position during instrumentation, which enables optimal alignment and spontaneous correction during surgery. The angle correction in the VP group primarily results from stabilizing the affected vertebra and pain reduction, thereby allowing patients to maintain an erect posture. Both groups experienced a significant loss of sagittal angle correction after 1 year, with the Cobb angle at 1 year being larger than immediately postoperatively (Fig. 3). Maintaining sagittal angle correction over time is difficult due to potential issues, such as screw loosening, which has a reported rate of 22.5% (95% CI, 10.8–36.6), and screw migration, which occurs at a rate of 1.3% (95% CI, 0.1–3.5) [17]. These screw-related problems cause a collapse of the instrumented vertebral body and increased biomechanical stress on adjacent vertebrae, resulting in a loss of correction. The progression of kyphotic angle in the VP group may be driven by fractures at adjacent levels or ongoing collapse of the operated vertebra. A previous study by Gu et al. [18] revealed that hybrid stabilization maintained vertebral alignment more effectively than balloon kyphoplasty; however, both groups demonstrated a similar trend of losing local angle correction, consistent with the results in our study. Multiple studies have revealed the effectiveness of VP and kyphoplasty in correcting the kyphotic and wedge angles. Gamal et al. [19] compared both procedures and revealed that VP reduced the kyphotic angle from 9.18° to 6.0° and increased the mean vertebral height from 63.62%±3.71% to 73.15%±1.41%. Moreover, Wang et al. [20] demonstrated that VP improved both the wedge and kyphotic angles. The loss of correction emphasizes the challenge of maintaining long-term vertebral alignment in osteoporotic fractures, likely due to compromised bone quality, biomechanical stress on adjacent vertebrae, patients with severe comorbidities, such as Parkinson’s disease, rheumatoid arthritis, and those who required substantial correction due to pronounced preoperative local kyphosis and vertebral collapse with a significant loss of correction [21].

Fig. 3

Local Cobb angle of preoperative, postoperative, and 1-year postoperative of both procedures. Sagittal angle correction of both procedures. SS, short-segment instrumentation; VP, vertebroplasty.

Previous studies revealed higher complication rates for surgical treatments, with Spiegl et al. [16] noting a 13%–16% rate [10,11], including implant failures and infections. Our results are congruent with these studies, indicating that both VP and SS+VP experienced a significant loss of correction >1 year despite initial benefits. This emphasizes the importance of balancing the immediate advantages of surgical intervention against the risks of complications and long-term maintenance.

However, our study’s result demonstrates some contrasting information. The number of reoperation and complication rates were higher in the SS+VP group, although not statistically significant, and may still be clinically relevant. These results, which contrasted with the favorable 1-year clinical outcomes in the SS+VP group, indicating interventions after the primary operation (one extended fusion 1 month after primary surgery and two wound debridement), did not adversely affect patients’ quality of life at the 1-year follow-up. The increased incidence of progressive kyphosis and adjacent fractures in the SS+VP group could be attributed to the higher biomechanical demands placed on adjacent segments by the instrumentation. These radiographic results were evident, but they did not significantly impair the patients’ reported outcomes, as both groups demonstrated improved quality-of-life scores at 1 year. Furthermore, the adjacent fractures that occurred in both groups were conservatively treated, with patients experiencing only minimal pain, indicating a limited effect on overall clinical outcomes.

The study has several limitations, including its retrospective design and the relatively small sample size, which have contributed to the large percentage of variations observed in perioperative details and radiographic outcomes. This limitation indicates that with a larger cohort, these trends could potentially reach statistical significance, thereby warranting further investigation. The use of propensity-score matching helps mitigate some biases, but prospective randomized controlled trials are warranted to confirm these results. Additionally, the 1-year follow-up period may be insufficient to capture long-term complications, such as proximal junctional kyphosis and proximal junctional failure, which appears over time, particularly in cases involving posterior fixation. Future studies with extended follow-up periods are warranted to comprehensively assess the durability of outcomes and monitor for late-onset complications. Collecting further long-term data would provide a clearer picture of the sustainability of sagittal alignment and functional outcomes related to each treatment method.

Conclusions

SS+VP is a more effective treatment for osteoporotic vertebral fractures in terms of radiographic correction and quality of life improvements, despite the higher intraoperative risks. These results provide crucial information for clinicians in making informed decisions about the most appropriate surgical interventions for OVFs, thereby emphasizing the need for individualized treatment plans to optimize patient outcomes. Further research is warranted to confirm these results in larger cohorts and over longer follow-up periods.

Article information Continued