Risk factors of residual back pain after vertebral augmentation in osteoporotic vertebral compression fracture patients: a systematic review and meta-analysis

Abstract

Background

Residual back pain (RBP) following vertebral augmentation negatively impacts clinical satisfaction and compromises both the physical and psychological well-being of affected patients. This meta-analysis aimed to identify risk factors associated with RBP after vertebral augmentation in patients with osteoporotic vertebral compression fractures (OVCF).

Methods

We searched literature in the PubMed, Embase, Web of Science and Cochrane Library. PRISMA guidelines were followed in this review. The Newcastle–Ottawa Scale (NOS) were used to evaluate the quality of included studies. Odds ratio (OR), mean difference (MD) and 95% confidence interval (CI) of potential risk factors were calculated via RevMan5.4.

Results

A total of 11 studies and 3290 patients were included in this meta-analysis. 12 risk factors were assessed and the outcome showed that lower BMD (MD = 0.39, 95% CI 0.27 ~ 0.52, p < 0.01), lower BMI (MD = -1.22, 95% CI -2.38 ~ -0.06, p = 0.04), depression (OR = 2.72, 95% CI 1.48 ~ 5.01, p < 0.01), adjacent vertebral fracture (OR = 2.53, 95% CI 1.42 ~ 4.50, p < 0.01), posterior fascia injury (OR = 3.94, 95% CI 3.10 ~ 5.00, p < 0.01), intravertebral vacuum cleft (OR = 2.33, 95% CI 1.56 ~ 3.50, p < 0.01), severe paraspinal muscle degeneration (OR = 6.25, 95% CI 4.09 ~ 9.53, p < 0.01), facet joint violation (OR = 7.71, 95% CI 3.50 ~ 17.00, p < 0.01), unsatisfied bone cement distribution (OR = 2.82, 95% CI 1.67 ~ 4.76, p < 0.01), less bone cement volume (MD = -0.24, 95% CI -0.45 ~ -0.03, p = 0.02), less recovery rate of anterior vertebral height (MD = -3.46, 95% CI -6.22 ~ -6.09, p = 0.01), less postoperative local kyphosis correction rate (MD = -4.74, 95% CI -6.43 ~ -3.06, p < 0.01) were associated with postoperative RBP in OVCF patients.

Conclusion

Lower BMD, lower BMI, depression, adjacent vertebral fracture, posterior fascia injury, intravertebral vacuum cleft, severe paraspinal muscle degeneration, facet joint violation, unsatisfied bone cement distribution, less bone cement volume, less recovery rate of anterior vertebral height, and less postoperative local kyphosis correction rate were associated with an increased risk of RBP after vertebral augmentation in OVCF patients. Early identification and targeted management of high-risk patients may help reduce RBP incidence.

Introduction

Osteoporotic vertebral compression fracture (OVCF) is one of the most common and serious complications of osteoporosis in the elderly [1, 2]. It often causes severe back or lower back pain and restricted spinal mobility, significantly impairing the quality of life in older adults. For OVCF patients without neurological signs or symptoms, conservative treatments are firstly recommended, including bed rest, brace support, and analgesic medication [3]. Vertebral augmentation is recommended for patients suffering intractable pain unresponsive to conservative treatments, among which percutaneous vertebroplasty (PVP) and percutaneous kyphoplasty (PKP) are two common and effective methods for elder patients in relieving fracture-related pain immediately with less risks of anesthesia and providing adequate stabilization at fractured site through minimally invasive wound [4–6].

Residual back pain (RBP) is defined as a postoperative visual analogue scale (VAS) score ≥ 4, based on previous literature in the absence of established guidelines [7]. RBP is considered an early postoperative complication, with a reported prevalence ranging from 5 to 20% [1, 7]. Unsatisfactory back pain negatively affects clinical satisfaction rate, as well as the physical and psychological health of OVCF patients [1, 5]. A number of studies have attempted to identify risk factors for RBP following vertebral augmentation; however, the exact contributing factors remain controversial.

The purpose of this meta-analysis was to identify potential risk factors associated with RBP in OVCF patients following PKP or PVP.

Materials and methods

Search strategy

A systematical literature review was conducted from PubMed, Embase, Web of Science, and Cochrane Library from the date of database inception to January 2024 according to the Preferred Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and protocol. We have registered this meta-analysis on PROSPERO with an ID of CRD42023461057. The searched key words were as follows: (“vertebral augmentation” OR “percutaneous vertebroplasty” OR “percutaneous kyphoplasty” OR “percutaneous augmentation”) AND (“residual pain” OR “residual back pain” OR “persistent pain” OR “postoperative pain”). No additional filters were used.

Eligibility and exclusion criteria

Selected studies meet following eligibility criteria were included: (1) Patients were diagnosed with OVCF and underwent PVP or PKP treatment; (2) RBP was defined as a postoperative VAS score ≥ 4; (3) Studies focusing on risk factors of RBP and involving over 50 participants.

Literature exclusion criteria included: (1) Case reports, meta-analyses, systematic reviews, comments, animal studies, meeting reports or letters, duplicates; (2) Patients diagnosed with vertebral fracture caused by tumor, infection, or tuberculosis; (3) Articles without available data or article data could not be extracted.

To improve the methodological quality and reliability of the pooled results, we included only studies with a sample size of more than 50 patients. This threshold was chosen to minimize the influence of small-study effects and to ensure more stable and generalizable estimates across the included literature.

Literature screening and data extraction

Two reviewers independently conducted literature search, screened the titles and abstracts of searched literature, and selected studies according to the uniform inclusion and exclusion criteria. If there is any discordance, a third researcher or a team discussion offer a solution.

The extracted data from qualified studies mainly included authors, years of publication, study subjects, number of patients, age, gender, treatment method (PVP or PKP), and reported risk factors in each article. This procedure was conducted by two reviewers.

Quality assessment

The Newcastle-Ottawa Scale (NOS) was applied to included literature to evaluate the risk of literature bias. The NOS score is used for non-randomized controlled studies, and the full score is 9. Studies NOS score over 6 are considered to be of high quality, and less than 6 being a low-quality study. Only studies with NOS scores ≥ 6 were included in this meta-analysis.

Two independent reviewers assessed the quality of evidence for the included studies according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) protocol. One study was assessed and assigned a rating of high, moderate, low, or very low, taking into account factors such as study design, limitations, and results.

Statistical analysis

Risk factors in literature associated with RBP are collected from at least two included studies. Dichotomous data are presented as odds ratios (OR) and 95% confidence interval (CI). Continuous data are evaluated using mean difference (MD) and 95% CI. Combined OR or MD with 95% CI are calculated for each risk factor separately. The results of risk factors will be represented as forest plots. The p < 0.05 is regarded as statistically significant. RevMan5.4 (The Cochrane Collaboration, Oxford, UK) is used to process all statistics. A random-effect model is used to calculate OR when I² > 50%; otherwise, a fixed-effect model is applied if I² < 50%. Sensitivity analysis was performed when studies showed strong heterogeneity. Publication bias was assessed with funnel plots.

Results

Included studies

A total of 440 records were identified by searching in PubMed, Embase, Web of Science, and the Cochrane library. 196 records were removed after duplicate checking. After screening titles and abstracts among the remaining 244 records, 218 records were excluded due to irrelevant articles, case reports, meeting reports, animal studies, reviews and so on. 26 articles were reviewed by full-text furtherly, and finally 11 studies were included for this meta-analysis [8–18]. A PRISMA flow diagram of included studies process was demonstrated in Fig. 1.

Fig. 1.

PRISMA flow diagram of included studies

Study characteristics

All included 11 articles were retrospective studies and were carried out in China. A total of 3290 participants were included in this meta-analysis, comprising 1191 males and 2099 females (Table 1). Four articles were published in Chinese, and the remaining 7 were in English. 5 studies rated 8 points according to the NOS score [10–12, 15, 18], another 5 studies scored 7 [8, 9, 11, 14, 17], and the rest one study got 6 points [16] (Table 2). The quality of evidence for each included studies was rated as very low based on the GRADE protocol.

Table 1.

Characteristics of included studies

Author/year	Country	Type of study	Treatment method	RBP group			Control group			Risk factors
				Sample (n)	Age (y)	Male/female	Sample (n)	Age (y)	Male/female
Yang 2019 [8]	China	Retrospective	PVP	60	69.38	19/41	60	69.19	21/39	Preoperative bone mineral density Lumbodorsal fascia contusion Number of fractures Cement volume injected per level Cement distribution Depression
Li 2020 [9]	China	Retrospective	PKP	52	76.2	21/31	163	74.5	51/112	Intravertebral vacuum cleft Posterior fascia oedema Facet joint violations Separated cement distribution
Chang 2021 [10]	China	Retrospective	PVP	61	69.04	23/38	61	68.77	27/34	Bone cement distribution Bone cement leakage Adjacent segment fracture Soft tissue injury Bone mineral density
Li 2021 [11]	China	Retrospective	PVP/PKP	37	-	7/30	231	-	46/185	Intravertebral vacuum cleft Posterior fascia oedema Severe paraspinal muscle degeneration Blocky cement distribution
Bo 2022 [12]	China	Retrospective	PVP	56	70.2	21/35	100	69.5	33/67	Poor sagittal parameters Sarcopenia
Chen 2022 [13]	China	Retrospective	PVP	119	76.51	30/89	121	75.71	29/92	Preoperative thoracolumbar fascial injury Bilateral puncture Bone mineral density Amount of bone cement injection Anterior vertebral height recovery rate
Liu 2022 [14]	China	Retrospective	PVP/PKP	33	71.60	13/20	184	71.08	75/109	Body mass index Lumbodorsal fascia injury Bone mineral density Adjacent vertebral fractures Postoperative vertebral height recovery rate Improvement rate of postoperative Cobb angle
Gao 2023 [15]	China	Retrospective	PVP/PKP	86	77.4	34/52	790	75.9	275/515	Posterior fascia injury Paraspinal muscle fatty degeneration Facet joint violation
Liu 2023 [16]	China	Retrospective	PVP	24	-	-	194	-	-	Bone cement distribution
Wang 2023a [17]	China	Retrospective	PVP/PKP	28	78.54	15/13	155	78.83	67/78	Body mass index Bone mineral density Soft tissue injury of low back Cause of injury Past history of low back pain Short-term complications Fracture nonunion Recurrent fracture Postoperative vertebral height recovery rate Improvement rate of sagittal Cobb angle Total diffusion coefficient of bone cement
Wang 2023b [18]	China	Retrospective	PVP	46	78.2	20/26	629	75.7	264/365	Low pre-bone mineral density Multiple vertebral fractures Posterior fascia injury Cement diffusion volume rate < 0.2 Facet joint violation Depression

Table 2.

Quality assessment of included studies by the Newcastle-Ottawa scale

Study	Selection				Comparability		Outcome			Total
	Representativeness of the exposed cohort	Selection of the nonexposed cohort	Ascertainment of exposure	Demonstration that outcome of interest was not present at the start of study	Study controls for age, sex, marital status	Study controls for any additional factor	Assessment of outcome	Was follow up long enough for outcomes to occur	Adequacy of follow-up of cohorts
Yang 2019 [8]	1	1	1	1	1	0	1	0	1	7
Li 2020 [9]	1	1	1	1	1	0	1	0	1	7
Chang 2021 [10]	1	1	1	1	1	0	1	1	1	8
Li 2021 [11]	1	1	1	1	1	0	1	1	1	8
Bo 2022 [12]	1	1	1	1	1	0	1	1	1	8
Chen 2022 [13]	1	1	1	1	1	0	1	0	1	7
Liu 2022 [14]	1	1	1	1	1	0	1	0	1	7
Gao 2023 [15]	1	1	1	1	1	0	1	1	1	8
Liu 2023 [16]	1	1	1	1	1	0	1	0	0	6
Wang 2023 a [17]	1	1	1	1	1	0	1	0	1	7
Wang 2023 b [18]	1	1	1	1	1	0	1	1	1	8

Meta-analysis results

17 risk factors were reported across the 11 included studies, and we categorized these reported risk factors into 4 main domains as follows: patients characteristics, preoperative radiological characteristics, surgical factors, and postoperative radiological parameters. Patient characteristics included bone mineral density (BMD), body mass index (BMI), and depression. Preoperative radiological characteristics involved adjacent vertebral fracture, multiple-segment fracture (n ≥ 2), posterior fascia injury, intravertebral vacuum cleft, and severe paraspinal muscle degeneration. Surgical factors contained facet joint violation, separated cement distribution, bone cement leakage, unsatisfied cement distribution, bone cement volume, and bilateral puncture. Postoperative radiological parameters comprised anterior vertebral height recovery rate, local kyphosis correction degree, and local kyphosis correction rate.

Patient characteristics

BMD presented by T score was available in 10 studies [8–15, 17, 18], and the pooled data proved that lower BMD was associated with RBP [MD = 0.39, 95% CI (0.27, 0.52), p < 0.01]. A random effect model was used on account of strong heterogeneity (I² = 88%, p < 0.01).

Six included articles reported BMI and the pooled MD was − 1.22 kg/m² [95% CI (−2.38, −0.06), p = 0.04], suggesting that lower BMI was a risk factor of RBP [9, 11, 12, 14, 15, 17]. There was a strong heterogeneity among the 6 studies (I² = 89%, p < 0.01).

Additionally, two studies suggested that OVCF patients with comorbid depression were more likely to develop RBP with no heterogeneity [OR = 2.72, 95% CI (1.48, 5.01), p < 0.01, I² = 0%] [8, 18]. Forest plots of patient characteristics were shown in Fig. 2.

Fig. 2.

Forest map of the relationship between RBP and factors of patient characteristics. A BMD (bone mineral density). B BMI (body mass index). C Depression. RBP, residual back pain

Preoperative radiological characteristics

Two studies reported adjacent vertebral fracture, and the pooled OR was 2.53 [95% CI (1.42, 4.50), p < 0.01] showed adjacent vertebral fracture was related to RBP without heterogeneity (I² = 0%, p = 0.97) [10, 14].

Multiple-segment fracture (n ≥ 2) data were extracted from 3 studies, but the result demonstrated that multiple-segment fracture was not correlated with RBP [OR = 1.33, 95% CI (0.33, 5.31), p = 0.68, I² = 90%] [9, 10, 18].

Meta-analysis of 9 studies conducted by a fixed effect model reported that posterior fascia injury had an effect on RBP after vertebral augmentation [OR = 3.94, 95% CI (3.10, 5.00), p < 0.01]. There was low heterogeneity occurred between the 9 studies (I² = 22%, p = 0.25) [8–11, 13–15, 17, 18].

A total of 3 included studies revealed a relationship between intravertebral vacuum cleft and postoperative RBP [OR = 2.33, 95% CI (1.56, 3.50), p < 0.01]. A fixed effect model was used to perform the meta-analysis and no heterogeneity existed among the three articles (I² = 0%, p = 0.67) [9, 11, 15].

Paraspinal muscle fatty degeneration is categorized into 0–1, 2, and 3–4 grade according to Goutallier grade classification [11]. Goutallier grade 3–4 was considered as severe paraspinal muscle degeneration [19]. 2 studies demonstrated that severe muscle fatty degeneration (Goutallier grade 3–4) was associated with postoperative residual pain following vertebral augmentation by a fixed effect model [OR = 6.25, 95% CI (4.09, 9.53), p < 0.01] [11, 15]. There was no heterogeneity among the included two studies (I² = 0%, p = 0.44). Forest plots of preoperative radiological characteristics were shown in Fig. 3.

Fig. 3.

Forest map of the relationship between RBP and factors of preoperative radiological characteristics. A Adjacent vertebral fracture. B Multiple-segment fracture. C Posterior fascia injury. D Intravertebral vacuum cleft. E Paraspinal muscle fatty degeneration. RBP, residual back pain

Surgical factors

Facet joint violation was defined as the facet joint invaded by the needle trajectory during the surgery procedure, which could be distinguished by postoperative axial computed tomography images [9]. Meta-analysis of 3 studies using a random effect model (I² = 64%, p = 0.06) showed that facet joint violation was a risk factor associated with RBP after PKP or PVP in OVCF patients [OR = 7.71, 95% CI (3.50, 17.00), p < 0.01] [9, 15, 18].

Separated cement distribution was identified by postoperative X-Ray radiographs with obvious separated cement masses without contact. 2 included studies presented that separated cement distribution was not a risk of RBP in OVCF patients underwent PKP or PVP [OR = 0.93, 95% CI (0.15, 5.71), p = 0.93] [9, 11]. A random effect model was used on account of strong heterogeneity (I² = 92%, p < 0.01).

Pooled data on bone cement leakage were extracted from 6 studies, and there was no obvious relationship between bone cement leakage and RBP [OR = 1.82, 95% CI (0.96, 3.43), p = 0.06, I² = 69%] [9–11, 14, 15, 18].

Four studies mentioned unsatisfied bone cement distribution [8, 10, 15, 16], and the pooled OR was 2.82 [95% CI (1.67, 4.76), p < 0.01], suggesting unsatisfied bone cement distribution was a risk factor of RBP. There was a moderate heterogeneity among the 4 studies (I² = 52%, p = 0.10).

A total of 10 studies explored the relationship between bone cement volume and RBP [8–15, 17, 18]. The pooled MD was − 0.24 ml with 95% CI (−0.45, −0.03) (p = 0.02) showed less bone cement volume was significantly associated with RBP. A random effect model was used on account of strong heterogeneity (I² = 86%, p < 0.01).

In terms of method of puncture, pooled data from 6 studies suggested that bilateral puncture was irrelative with postoperative RBP [OR = 1.02, 95% CI (0.63, 1.65), p = 0.93, I² = 70%]. Forest plots of surgical factors were shown in Fig. 4.

Fig. 4.

Forest map of the relationship between RBP and factors of surgical factors. A Facet joint violation. B Separated cement distribution. C Bone cement leakage. D Unsatisfied bone cement distribution. E Bone cement volume. F Bilateral puncture. RBP, residual back pain

Postoperative radiological parameters

6 studies involved postoperative anterior vertebral height recovery rate [9, 11, 13, 15, 17, 18], and meta-analysis of these studies using a random effect model (I² = 94%, p < 0.01) showed that less recovery rate of anterior vertebral height was a risk factor associated with RBP [MD = −3.46%, 95% CI (−6.22, −6.09), p = 0.01].

As for the local kyphosis improvement, pooled data from 2 studies suggested less local kyphosis correction degree was not related with RBP [MD = −0.17°, 95% CI (−0.80, 0.47), p = 0.61, I² = 0%] [11, 18]. However, other 2 included studies demonstrated less postoperative local kyphosis correction rate was a risk factor for RBP without heterogeneity [MD = −4.74%, 95% CI (−6.43, −3.06), p < 0.01, I² = 0%] [14, 17]. Forest plots of postoperative radiological parameters were shown in Fig. 5.

Fig. 5.

Forest map of the relationship between RBP and factors of postoperative radiological parameters. A Postoperative anterior vertebral height recovery rate. B Local kyphosis correction degree. C Postoperative local kyphosis correction rate. RBP, residual back pain

Sensitivity and publication bias analyses

Sensitivity analyses were performed to evaluate the robustness of our findings. In the 10 studies related to BMD, excluding one study at a time did not significantly impact the pooled SMDs, indicating the stability and reliability of BMD. We did not conduct the sensitivity analyses on other factors due to the limited literatures.

Publication bias was assessed with funnel plots, and the funnel plot of posterior fascia injury is symmetrical (Fig. 6).

Fig. 6.

Funnel plot of posterior fascia injury. The x-axis indicates OR of posterior fascia injury, and the y-axis indicates SE (log[OR]). OR, odds ratio. SE, standard error

Discussion

Compared with traditional open reduction and internal fixation for OVCF in elder patients without neurologic signs or symptoms, vertebral augmentation obviously shortens operation time and hospital stay, eliminates risks of general anesthesia, provides early ambulation in hours postoperatively, and prevents long-term bed-rest related complications such as bedsore, deep venous thrombosis, worsening osteoporosis, pulmonary embolism [3]. Two high-quality randomized controlled trials revealing that PVP was not superior to conservative treatment in relieving pain at 1, 3, or 6 months postoperatively in elderly OVCF patients [20, 21]. About 5–20% of patients still report intolerable RBP after surgery, negatively affecting their quality of life and delaying discharge [1, 22]. To figure out risk factor associated with RBP following vertebral augmentation in OVCF patients, this meta-analysis was conducted and indicated that lower BMD, lower BMI, depression, adjacent vertebral fractures, posterior fascia injury, intravertebral vacuum cleft, severe paraspinal muscle degeneration, facet joint violation, unsatisfied bone cement distribution, less bone cement volume, less recovery rate of anterior vertebral height and less postoperative local kyphosis correction rate were all significantly associated with postoperative RBP.

Patient characteristics

Lower BMD T-score, reflecting reduced calcium and mineral content in bone, indicates more severe osteoporosis. Some studies pointed lower BMD is a risk factor of vertebrate re-fracture affect vertebral augmentation procedure [23, 24]. New symptomatic vertebral compression fractures or trabecular micro fractures in areas lacking cement support may explain RBP in lower BMD patients. In addition, lower BMI was regarded as a risk factor of RBP by this meta-analysis. Lower BMI is related with increased bone loss in postmenopausal women, resulting lower BMD [25]. Also, lower BMI is associated with new vertebral compression fractures [26]. Furthermore, psychological disorders such as depression and anxiety have been shown to worsen postoperative pain and increase the need for opioid prescriptions [27–29]. Therefore, comprehensive perioperative management for OVCF patients should include anti-osteoporosis treatment, psychological support, and nutritional optimization.

Preoperative radiological characteristics

Local kyphosis and sagittal imbalance of global spine caused by vertebral collapse are considered as major contributors to low back pain [30]. Adjacent vertebral fracture and multiple-segment fractures may exacerbate these spinal deformities. In this meta-analysis, adjacent vertebral fracture was proved to be a risk factor for RBP, whereas multiple-segment fracture was not. Extracted data from included studies to calculate outcome of the two potential risk factors were different, thus more high-quality studies are needed. Notably, some OVCF patients may suffer RBP caused by newly unrecognized adjacent vertebral fractures in the early stage following surgery.

Posterior fascia injury was another important risk factor of postoperative RBP [8, 9, 11, 15, 17]. Yan et al. [31], in a prospective cohort study, reported that patients with thoracolumbar fascia injury experienced significantly more pain and disability after PVP compared to those without fascia injury, highlighting a correlation between thoracolumbar fascia injury and residual back pain. Luo et al. [32] obtained similar results that postoperative RBP was associated with injured thoracolumbar fascia and revealed low BMI, hypertension and sarcopenia were risk factors for thoracolumbar fascia injury. Posterior fascia injury in elder OVCF patients usually occurred at more than one level, and more severe injured thoracolumbar fascia caused by worse trauma was concomitant with unsatisfactory pain relief after a PKP [33]. To address RBP after vertebral augmentation related to posterior fascia injury, the thoracolumbar interfascial block with local anesthesia was performed during PKP and proved to be an effective method in minimizing RBP and reducing perioperative rescue analgesic use [34].

Intravertebral vacuum cleft was probably caused by bone trabecula ischemia secondary to trauma and presented a radiological sign of an intravertebral transverse or linear radiolucent shadow (Fig. 7) [35]. PVP was found to be an effective treatment in providing pain relief and stabilization in OVCF patients with intravertebral vacuum cleft [36]. However, the presence of intravertebral vacuum cleft was a potential risk of progressive vertebral collapse, delayed pain relief and diminished improvement in quality of life in OVCF patients after percutaneous vertebral augmentation [37, 38]. The mechanism of intravertebral vacuum cleft causing postoperative RBP is not clear. Researches pointed out that intravertebral vacuum cleft in compressed vertebra may result in nonunion of cement, therefore the unhealed bone-cement interface causing dynamic mobility probably contributed to intractable RBP [39, 40]. Qi et al. [41] found that intravertebral vacuum cleft was related to intravertebral bone cement displacement in OVCF patients treated by PKP, which was considered as another element causing RBP.

Fig. 7.

Intravertebral vacuum cleft in a thoracic vertebra was visible on axial view of computed tomography

Paraspinal muscle degeneration has been reported to be associated with low back pain in lumbar spinal stenosis patients, and paraspinal muscle rehabilitation accounted for alleviating low back pain [42]. According to Goutallier grade classification, grade 3–4 represent severe muscle degeneration. However, studies about the impact of severe muscle degeneration on RBP after PKP or PVP was rare. Jeon et al. [43] reported that paraspinal muscle fatty degeneration was strongly associated with progressive vertebral body collapse in OVCF patients. Based on that bone cement was unable to distribute fully in fractured vertebral body, micro progressive collapse may occur in undistributed area and subsequently lead to RBP after vertebral augmentation. Gao et al. [15] also referred to functional cross-sectional area of the paraspinal muscle as a potential factor associated with RBP, though insufficient data precluded further analysis. Preoperative assessment and rehabilitation of paraspinal muscles are essential components of care.

Surgical factors

Facet joint violation was rarely reported in previous literature about percutaneous vertebral augmentation. During lumbar pedicle screw placement in open or minimally invasive techniques, facet joint violated by pedicle screw has been associated with adjacent segment degeneration, the need for reoperation, and postoperative back pain [44, 45]. Li et al. [46] reported an 18.9% incidence of facet joint violation by puncture trocar during PKP procedure in OVCF patients. Since mechanoreceptors and nociceptors are attached to facet joint capsules and cartilage, direct neurons impairment and inflammatory reaction second to injury may be the possible reasons of RBP following PKP in zygapophyseal joint injured patients [9, 15, 46]. Anatomical features as large facet joint angle and narrow pedicle diameter added the risk of the entry point closer to zygapophyseal joint [46]. Facet joint violation by puncture trocar is categorized as one of iatrogenic injures, which should be avoided regardless of its relevance to RBP.

As for distribution of bone cement, satisfied cement distribution was defined as the spread of cement from the superior to the inferior end plate, extending from the medial cortex of the ipsilateral pedicle to the medial cortex of the contralateral pedicle, and reaching from the anterior cortex of the vertebral body to the posterior third of the vertebral body [8]. Satisfied spread cement was able to provide sufficient and symmetrical support between the superior and inferior endplates. Unsatisfied cement distribution means lack of cement at the broken area in a collapsed vertebrate, thus micromovements of fractured trabecula produce persistent pain when postoperative patients ambulance. Another cement distribution evaluation method is whether the cement spread was confluent. Confluent cement distribution is considered to be associated with curative effects, and provide earlier pain relief and rehabilitation than separated distribution [47]. However, this meta-analysis revealed that separated cement distribution was not a factor for RBP following vertebral augmentation. Separated cement masses symmetrically distributing on both sides of the vertebral body are able to stabilize micromovements in a fractured vertebrate.

Studies have indicated that sufficient bone cement diffusion and cement volume were related with pain alleviation [48, 49]. Our research showed less bone cement volume was associated with RBP. Firstly, a small amount bone cement may not ensure optimal cement distribution within the vertebral body. Secondly, the limited vertebral height restoration restricts the amount of injected bone cement [48]. A smaller volume of injected cement reflects inadequate correction of sagittal alignment, which is also linked to RBP [30].

Postoperative radiological parameters

Postoperative radiological parameters such as C7-sagittal vertical axis (SVA), lumbar lordosis (LL), pelvic tilt (PT), sacral slope (SS), pelvic incidence (PI), and vertebral body kyphosis malformation and height loss are critical in evaluating outcomes. Only one included study investigated relationship between SVA, LL, PT, SS, PI and RBP and indicated larger C7-SVA and PI-LL mismatch were associated with of RBP [12]. Several studies have reported vertebral augmentation not only provides pain relief and improves quality of life, but can also correct local kyphosis and restore vertebral body height [50–52]. Slight segmental kyphosis degree was associated better pain improvement after vertebroplasty in OVCF patients [53]. Restoration of anterior vertebral height is essential for achieving adequate cement volume and support. Lumbar degenerative diseases patients with sagittal imbalance had poorer outcomes in terms of quality of life after lumbar decompression with a short fusion [54]. Our research showed less recovery rate of anterior vertebral height and less postoperative local kyphosis correction rate were significant factors associated with RBP. However, local kyphosis correction degree was not an identified risk factor according the pooled OR and 95% CI. This may be attributed to the limited number of studies available for data synthesis.

Several risk factors in our meta-analysis, including BMD (I² = 88%), BMI (I² = 89%), multiple-segment fracture (I² = 90%), separated cement distribution (I² = 92%), bone cement volume (I² = 86%), and anterior vertebral height recovery rate (I² = 94%) demonstrated substantial heterogeneity. This may stem from multiple sources. First, differences in patient demographics such as sex ratio and baseline osteoporosis severity could have influenced these outcomes. Postmenopausal women tend to have more severe osteoporosis than men. Additionally, in some patients with severe baseline osteoporosis, the degree of vertebral compression is more pronounced, leading to more intense fracture-related pain. As a result, the improvement in pain following vertebral augmentation may be less significant in these individuals. Second, variations in vertebral augmentation techniques may contribute to the observed heterogeneity. Compared to PVP, PKP has demonstrated superior correction of the Cobb angle and greater restoration of anterior vertebral body height, which may result in better pain relief following the procedure [55]. Additionally, patients with OVCF who received high-viscosity bone cement during PVP or PKP experienced a more significant reduction in VAS score than those treated with low-viscosity cement [56]. Furthermore, the follow-up durations among the included studies were inconsistent. Longer follow-up periods may influence the detection rate of residual back pain and the identification of associated risk factors. These factors collectively highlight the need for future studies with standardized methodologies to reduce heterogeneity and improve comparability. Some included studies enrolled patients treated with either PVP or PKP; however, the exact number of RBP cases specific to each procedure was not reported. Additionally, data on elderly patients (age > 65 years) and those with severe osteoporosis (BMD T score < −3.5) were insufficient, precluding further subgroup analysis.

Based on the identified risk factors for RBP after vertebral augmentation in OVCF patients—including lower BMD, lower BMI, depression, adjacent vertebral fracture, posterior fascia injury, intravertebral vacuum cleft, severe paraspinal muscle degeneration, facet joint violation, unsatisfactory bone cement distribution, lower bone cement volume, reduced recovery rate of anterior vertebral height, and less postoperative local kyphosis correction—several clinical recommendations can be made. Patients with low BMD or BMI should be evaluated for comprehensive osteoporosis management, including pharmacologic treatment as teriparatide and nutritional support. Screening and managing depression preoperatively may improve patient outcomes and reduce pain perception after surgery. Adequate bone cement volume and optimal distribution should be ensured during surgical procedure to enhance postoperative vertebral stability, which may contribute to reduce persistent pain. Surgeons should aim for satisfactory restoration of anterior vertebral height and correction of local kyphosis, as these factors are associated with improved pain relief and spinal alignment. Special care should be taken to avoid facet joint violation during needle placement. Thoracolumbar interfascial block with local anesthesia is recommended for OVCF patients with posterior fascia injury, as it may help improve postoperative VAS scores. Patients with intravertebral vacuum cleft or adjacent vertebral fractures should be closely monitored, as these conditions are linked with a higher risk of RBP. Early rehabilitation programs focused on strengthening the paraspinal muscles may help mitigate the effects of muscle degeneration and support spinal mechanics. Additionally, other approaches to reduce residual pain have been implemented in clinical practice and have yielded satisfactory outcomes. Preoperative administration of zoledronic acid, medial branch block with local anesthetic, epidural steroid injections, and even repeat percutaneous vertebroplasty have been shown to be effective in reducing RBP in OVCF patients underwent vertebral augmentation [57–60].

Currently, there is a paucity of clinical guidelines specifically addressing RBP following vertebral augmentation. A recent guideline summary indicates that the available evidence is insufficient to confirm or refute an association between intravertebral vacuum cleft, posterior fascia edema, and RBP in OVCF patients after vertebral augmentation [61]. Given that most existing studies are retrospective in nature, high-quality, multicenter prospective studies are warranted to validate these potential risk factors and inform potential modifications to surgical guidelines.

Limitations

There are several limitations in this study. Firstly, all of included literature were retrospective studies in nature and rated as very low quality based on the GRADE protocol, which may limit the reliability and validity of the pooled estimates. Thus, more randomized controlled trials with high-quality and large sample are needed. Secondly, some potential risk factors as SVA and PI-LL were not analyzed due to lack of raw data. Thirdly, all included studies were conducted in China, which may limit the generalizability of our findings to broader or non-Chinese populations. Differences in healthcare systems, osteoporosis management protocols, surgical techniques, and cultural perceptions of pain may influence both the diagnosis and reporting of RBP. The indications for surgical intervention, surgical experience and postoperative rehabilitation may differ across countries. Moreover, ethnic and genetic factors may also contribute to variations in bone quality and pain perception. Therefore, caution is advised when extrapolating these results to other populations, and further high-quality, multicenter studies from different countries and regions are needed to validate these findings in a more globally representative cohort. Lastly, the definition of RBP currently lacks a standardized criterion. While most studies define RBP as a postoperative VAS score ≥ 4, the timing of postoperative pain assessment varies across studies. One included study [13] used a higher threshold, defining RBP as a postoperative VAS score ≥ 5, which may lead to fewer identified cases. In contrast, applying a lower threshold could overestimate the prevalence. These inconsistencies highlight the need for standardized RBP criteria in future research to ensure consistency and comparability across studies.

Conclusion

Lower BMD, lower BMI, depression, adjacent vertebral fracture, posterior fascia injury, intravertebral vacuum cleft, severe paraspinal muscle degeneration, facet joint violation, unsatisfied bone cement distribution, less bone cement volume, less recovery rate of anterior vertebral height, and less postoperative local kyphosis correction rate were identified as risk factors for RBP following vertebral augmentation in OVCF patients. In clinical practice, early identification of patients with these risk factors and implementation of appropriate preventive strategies are essential to reduce the incidence of RBP. However, as all included studies were rated as very low quality according to the GRADE protocol, these findings should be interpreted with caution. To better prevent RBP and achieve sustained pain relief and improved quality of life in OVCF patients, high-quality prospective studies and further research into the underlying mechanisms are warranted.

Abbreviations

RBP

Residual back pain

OVCF

Osteoporotic vertebral compression fracture

PVP

Percutaneous vertebroplasty

PKP

Percutaneous kyphoplasty

VAS

Visual analogue scale

NOS

the Newcastle–Ottawa Scale

OR

Odds ratio

MD

Mean difference

CI

Confidence interval

BMD

Bone mineral density

BMI

Body mass index

SVA

C7-sagittal vertical axis

LL

Lumbar lordosis

PT

Pelvic tilt

SS

Sacral slope

PI

Pelvic incidence

Authors’ contributions

H Wang and H Zhang: Writing– original draft, Methodology, Formal analysis, Conceptualization. C Xiao and K Zhang: Writing– review & editing. L Qi: Writing– review & editing, Visualization, Supervision, Methodology, Conceptualization.

Funding

The study was not supported by any funds.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.