Abstract
Study Design
Systematic review.
Objective
Intracardiac cement embolism (ICE), a rare but life-threatening complication of vertebral augmentation (VA), is poorly characterized. Our review summarizes the incidence, clinical characteristics, treatment and outcome of this complication.
Methods
Medline, Embase, and Cochrane databases were systematically searched from inception to April 1, 2025. References were cross-checked to identify additional relevant articles. Included publications underwent quality assessment using the Case Report (CARE) Guidelines and the Joanna Briggs Institute (JBI) Critical Appraisal Checklist, with pooled analysis of incidence, clinical characteristics, treatment, and outcomes.
Results
4 case-series and 96 case reports (181 cases) were included. The overall reporting rate for each study was relatively high, but low reporting rates in 11 sub-items may introduce bias. Among 115 cases with epidemiological data, osteoporotic fractures predominated (n = 83), followed by other diseases (n = 32). The incidence was low, but the exact value was unclear. Common symptoms included dyspnea and chest pain, with onset typically within 48 h postoperatively, though delays of months to 5 years occurred. Diagnosis relied on chest imaging. Treatment was individualized, but most reports lacked objective evidence to guide decisions. Asymptomatic patients required monitoring, while symptomatic ones may need surgery intervention. The outcome was uneventful for all but 10 patients.
Conclusions
ICE was reported only in case report or case-series. A comparative analysis of pre- and postoperative chest X-ray is recommended for screening, whil
Systematic Review Registration
PROSPERO identifier CRD 42023455121
Keywords
Introduction
Bone cement is indispensable in vertebral augmentation (VA) procedures, including percutaneous vertebroplasty (PVP), percutaneous kyphoplasty (PKP) and cement-augmented pedicle screw instrumentation (CAPSI). All involved an injection of cement into the vertebral body. In PVP and PKP, cement provides mechanical stability, enabling rapid pain relief, vertebral height restoration and kyphosis correction. 1 Similarly, cement improves pedicle screw fixation and pullout strength during CAPSI. 2 As excellent effect, cement is widely used for osteoporosis-related spinal disorders, spinal metastases, and et al. However, cement leakage is an unavoidable problem. One major complication of cement leakage is pulmonary cement embolism (PCE), which has been extensively reported. PCE is caused by cement leakage into the paravertebral venous system, subsequently migrating into the azygos vein or vena cava, traversing the right cardiac cavities, and ultimately embolizing in the pulmonary arterial tree. 3 Although all PCE must first pass through the heart, and some cement fragments may remain within the heart cavities, intracardiac cement embolism (ICE) during VA has been poorly reported, as contrasted to PCE.
As a rare complication, ICE exhibits heterogeneous manifestations ranging from asymptomatic presentations to catastrophic cardiopulmonary symptoms manifesting as chest pain and dyspnea, acute respiratory distress syndrome, or death. Due to limited evidence, the incidence, clinical characteristics, management and outcomes remain poorly characterized. In this study, we conducted a comprehensive literature review to address these knowledge gaps.
Materials and Methods
This review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA) 4 and Assessing the Methodological Quality of Systematic Reviews (AMSTAR) Guidelines, 5 and was also prospectively registered on the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD 42023455121).
Search Strategy and Databases
A literature search was performed by 2 independent authors on the Medline (via Pubmed), Embase, and Cochrane databases from database inception to April 1, 2025. Rough search strategy: (“heart” OR “cardiac”) AND (“bone cements” OR “methylmethacrylates”) AND (“vertebroplasty” OR “kyphoplasty” OR “pedicle screw” OR “augmentation”).
Accurate search strategy: (Taking retrieval strategy of pubmed as an example). #1: ((((((((vertebroplasty[MeSH Terms]) OR (kyphoplasty[MeSH Terms])) OR (balloon vertebroplasty)) OR (vertebroplasty, balloon)) OR (PVA)) OR (PVP)) OR (PKP)) OR (VP)) OR (Augmentation) #2: ((((((((pedicle screws[MeSH Terms]) OR (pedicle)) OR (screws)) OR (pedicle screws)) OR (screw)) OR (screw s)) OR (screwed)) OR (screwing)) OR (screws) #3: (((((Heart[MeSH Terms]) OR (Cardiac)) OR (Intracardiac)) OR (Intraatrial)) OR (Intraventricular)) OR (cardiopulmonary) #4: (bone cement) OR (methylmethacrylates) #5: (#1 OR #2) AND #3 AND #4.
Inclusion and Exclusion Criteria
Studies reporting characteristics, treatment, and outcomes (with or without follow-up data) of ICE after PVP, PKP, or CAPSI, were included. When multiple articles reported the same patient, the most comprehensive or recent publication was selected. Only English language papers were screened.
Study Selection
Articles were downloaded from each database and duplicates were removed. Two reviewers independently screened the full-text against predefined inclusion/exclusion criteria. Manual searches of the references cited by these included articles were done to identify additional studies. Discrepancies were resolved by discussion or arbitration by a third reviewer.
Data Extraction and Analysis
The aim of this systematic review was to aggregate, summarize, and analyze the incidence, clinical characteristics, treatment, and outcome data on all reports about ICE after VA.
Quality Evaluation of the Reports
Only case-series studies6-9 and case reports10-105 were identified so that the Case Report (CARE) Guidelines 106 and the Joanna Briggs Institute (JBI) Critical Appraisal Checklist 107 were used to evaluate the quality of these reports (Supplemental Table S1). “Reported-Sufficient” means the item was fully reported; “Reported-Not sufficient” means the item was reported, but insufficiently; “Not reported” means the item was not reported; and “Not applicable” means the item could not be applied.
Two reviewers independently conducted this assessment. Disagreements were resolved by discussion and further disagreements were resolved by a third researcher.
Analysis of Quality Evaluation Results
After evaluation as described above, the number of “Reported-Sufficient”, “Reported-Not sufficient”, “Not reported” and “Not applicable” items in each study were first counted and converted into percentages to evaluate the quality of each study. And then, the number of studies in each of the 28 sub-items of the CARE guidelines and each of sub-items of the JBI critical appraisal checklist were evaluated and classified. The sub-items with a “Not reported + Not applicable” rate exceeding 50% were analyzed to determine whether improvement was required in the future.
Results
Search and Selection
The literature search initially yielded 159 articles. Following screening for predefined criteria, 72 articles were retained. Manual reference searches of these articles yielded 29 additional articles; after removing 1 duplicate, 28 were included. In total, 4 case-series6-9 and 96 case reports,10-105 providing data on 181 cases of ICE after VA, were identified. However, in the study of Fadili Hassani S et al,
6
detailed information was reported only for 6 symptomatic patients and the characteristics of the remaining 66 asymptomatic cases were not documented. As no randomized controlled trial was identified, methodological quality and meta-analysis were not assessed. The selection process was shown in Figure 1 and the excluded studies with reasons listed in Supplemental Table S2. Screening process chart
Quality Assessment as per Sub-items of the CARE Guidelines
Level of Quality of the Case Reports and Case Series (Supplemental Table S3)
On analyzing each study based on the evaluation criteria for sub-items of the CARE guidelines, it was determined that there were case reports classified as “Sufficient” and “Not sufficient” in the “Reported” section, with a maximum of 92.9%, a minimum of 21.4%, and an average of 62.9%. Overall, the reporting level was moderate.
Level of Quality according to the 28 Sub-items (Supplemental Table S4)
Sub-item 1, “Title” (79%); Sub-item 3a, “Introduction”, (54%); Sub-item 3c, “Conclusion”, (62%); Sub-item 6, “Clinical findings”, (63%); Sub-item 8b, “Diagnostic challenges”, (96%); Sub-item 8d, “Prognostic characteristics”, (73%); Sub-item 10b, “Important follow-up test results”, (64%); Sub-item 10c, “Intervention adherence and tolerability”, (86%); Sub-item 10d, “Adverse and unanticipated events”, (74%); Sub-item 12, “Patient perspective”, (100%) and Sub-item 13, “Informed consent”, (61%), were evaluated as “Not reported + Not applicable” in more than 50% of the selected studies.
Quality Assessment for Sub-items of the JBI Critical Appraisal Checklist
In the case series report of Bohl MA et al 7 and Kong M et al, 8 only 1 ICE case was reported so that the JBI Critical Appraisal Checklist for case report was used to evaluate the quality of these 2 reports.
Level of Quality of the Case Reports and Case Series
On analyzing the 98 reports7,8,10-105 based on the evaluation criteria of the sub-items of the JBI critical appraisal checklist, the report rates of case reports classified as “Reported Item” (Reported-Sufficient + Reported-Not Sufficient) were a maximum of 100%, minimum of 62.5%, and average value of 88.8%. The case reports classified as “Not reported + Not applicable Item” had a maximum report rate of 37.5%, minimum of 0%, and average value of 11.2% (Supplemental Table S5).
Analysis of the 2 case series6,9 revealed the report rates classified as “Reported Item” to be a maximum of 100%, minimum of 90%, and median of 95%. Those classified as “Not reported + Not applicable Item” had a maximum report rate of 10%, minimum of 0%, and average value of 5.0% (Supplemental Table S6).
Level of Quality According to the 8 Sub-Items for Case Reports and the 10 Sub-Items for Case Series
Item 1, Item 2 and Item 3 were reported in 100% of the case reports. The lowest reporting rate (29%) was for Item 7 (Supplemental Table S7).
On evaluating the case series according to the checklist criteria for each of 10 sub-items, only Item 10 was flawed (Supplemental Table S8).
Epidemiology
Excluding 66 asymptomatic patients without relevant information in the study of Fadili Hassani S et al, 6 the epidemiological investigation was conducted for 25 male, 86 female and 4 unknown gender patients.
83 patients were definitively diagnosed with osteoporotic fracture. Mean age at presentation was 70.68 years (range: 51-86). 56 were treated with PVP, 22 with PKP, and 5 with screw-rod fixation system with PVP for fracture vertebra or with cement screw channel augmentation.7,10-13
The primary disease of the remaining 32 patients were below: spondylolisthesis with osteoporosis (1 cases), 14 spinal metastasis (5 cases),6,9,15-17 glucocorticoid-induced osteoporotic fracture (5 cases),18-22 hematological disease (5 cases of multiple myeloma,6,23-26 1 case of Waldenström disease, 6 1 case of drepanocytosis, 6 1 case of lymphoma 6 ), lumbar spondylolisthesis (2 cases),27,28 lumbar spinal stenosis (2 cases),29,30 the first lumbar fracture caused by car accident (1 case), 31 lumbar radicular syndrome (1 case), 32 vertebral fracture caused by unknown reason (6 cases)33-38 and unknown reason (1 cases). 39 Mean age was 62.61 years (range: 28-78). PVP or PKP was used to treat patients with spinal metastasis, glucocorticoid-induced osteoporotic fracture, hematological disease and vertebral fracture caused by unknown reason (5 out of 6), while screw-rod fixation system with VA was used to treat remaining cases14,27-33 besides “unknown reason”. 39
Incidence of ICE after VA
The incidence of ICE after VA was uncertain. To date, only Fadili Hassani S et al 6 evaluated the ICE incidence during PVP. In this study, among 1512 consecutive patients (1854 PVP procedures), chest imaging detected ICEs in 72 patients (3.9% of all PVP procedures), with symptomatic manifestations observed in 6 patients (0.32% of total procedures). Barakat et al 9 reported 9 symptomatic ICE cases after PVP among 3941 patients at their institutions during the same period, yielding a symptomatic ICE incidence of 0.23%. In another case-series study, 7 all 34 patients undergoing prophylactic PVP during spinal reconstruction had computed tomography (CT) evidence of cement extravasation. However, only 1 patient exhibited cement in superior vena cava, right atrium, right ventricle, and bilateral pulmonary arteries, and distally in lungs. The incidence of ICE in this study was 2.94% (1/34).
It was fact that chest imaging and echocardiography after VA were not routinely performed. In our study, only 4 cases clearly indicated that ICE was discovered by routine postoperative examination,23,30,40,41 while 2 cases were detected during regular physical check-up42,43 and 7 cases were identified during the treatment of other disease.24,44-49 The majority of the remaining patients were diagnosed by relevant examination after noticing the cement leakage or the injection into the veins by mistake during surgery, 11 or after the patients developed significant cardiac and/or lung discomfort. Notably, ICE was misdiagnosed as calcified coronary artery in 2 cases due to motion artifacts on CT,18,19 while in another case, a cardiac foreign body visible on X-ray was directly neglected by the clinician. 27
Clinical Features
Dyspnea and chest pain were the most common symptoms, occurring either alone or concurrently in over 80 patients. Intraoperative arrhythmia6,16,20,50-53 and asymptomatic presentations6,8,24,41-43,45,46,54,55 were not rare either. Symptom onset time varied among patients, but most (>40) developed symptoms within 48 hours postoperatively. Cardiac murmur,56-58 irregular cardiac rhythm,15,19,44,48,51-53,59-63 fever,14,18,61 pneumothorax, 64 syncope 25 and shock6,12,65-68 were observed in cases where physical examination was documented. In the case of patent foramen ovale 13 and ostium secundum atrial septal defect, 21 cement can further migrated to the left heart chambers and systemic circulatory system, potentially causing acute traumatic mitral regurgitation and acute decompensated heart failure. 21 Laboratory findings may show an increased plasma d-dimer level,14,32,33,36,57,62,66,69-72 increased plasma cardiac troponins level,26,29,32,34,39,47,51,66,70,73 increased creatine phosphate kinase level,14,29 increased lactate dehydrogenase level14,29,66 and oxygen desaturation (low SpO2).14,31,61,67,72,74,75 Ultimately, almost all patients were diagnosed by chest imaging, except for 1 who was diagnosed by autopsy due to rapid disease progression and death within 3 hours of onset, 50 and 1 who was diagnosed by three-dimensional echocardiography alone. 76
Cardiac free wall perforation with pericardial effusion was clearly documented in 43 cases. The pericardium, diaphragm,73,77,78 lung,64,79 costal cartilage 18 and aortic wall 65 may be further perforated. In 1 case, the ICE nearly perforated the ventricular septum rather than the free wall. 56 Interestingly, in another case, linear cement fragments were discovered within the pericardial effusion without an identifiable myocardial penetration site. 40
Management and Outcome
The Selection of Treatment Option and Clinical Outcome
Among 75 asymptomatic patients, 66 were initially treated conservatively without anticoagulants,6,24,54 5 were treated conservatively with anticoagulants,6,41,45,55 1 underwent open-heart surgery, 46 and 3 underwent percutaneous endovascular snare retrieval.8,42,43 A 73-year-old woman was referred for open-heart surgery after failed anticoagulation because of the rapid cement migration. 55 During snare retrieval in 3 patients, cement fractured during capture or traction due to its fragile nature. Complete fragment removal was achieved in 1 patient, 43 partial removal in another, 8 and retrieval failed in the third. 42 Unretrieved fragments either adhered to the cardiac wall 42 or migrated into the pulmonary vessel8,42 without clinical consequences. All 75 asymptomatic patients were uneventful at last.
Among 100 symptomatic patients, 30 initially received conservative treatment. Of these, 21 were managed with anticoagulants, 8 without anticoagulants, and 1 under observation only. 54 During anticoagulation therapy, 4 patients30,40,69,80 developed pericardial effusion, resulting in a switch to open-heart surgery. Another 3 patients died from respiratory failure, 15 heart and respiratory failure due to atrial fibrillation and pneumonia 57 and multiple organ failure secondary to wound infection, 39 respectively. Additionally, Fadili Hassani et al 6 reported a conservatively managed patient (without anticoagulation) with moderate acute dyspnea who was found to have a lung tumor invading the mediastinum and right pulmonary artery. This patient died 10 days after PVP from an unknown cause.
57 symptomatic patients chose open-heart surgery as the preferred treatment. One patient experienced intraoperative cement fragmentation with subsequent pulmonary vascular migration, though without clinical consequence
13 symptomatic patients received percutaneous endovascular snare retrieval, which was successfully completed without any accidents in only 5 patients.9,10,47,84,85 Cement embolus broke into fragments in 4 cases22,34,70,86: 2 had complete removal,34,86 1 had partial removal, 22 and 1 retained fragments. 70 The residual cement migrated into the pulmonary vessel without clinical consequences.22,70 For 3 other cases, open venous cut-down was performed after retrieving the cement to the inferior vena cava 51 or common femoral vein by percutaneous snare technique,12,66 due to large cement size. One patient experienced failed endovascular procedure and was conversed to the open-heart surgery in the report of Schoechlin S and colleague, 77 but the cause of failure was not provided. Among all symptomatic patients treated with the percutaneous technique, only 1 died from metastatic lung cancer 3 months after the procedure. 9
In addition, 6 cases did not fit the above categories. An 80-year-old man was initially misdiagnosed as thromboembolic mass and treated with enoxaparin and oxygen. 74 Although open-heart surgery eventually confirmed the diagnosis, he still died 5 days later from septic shock. Lee V and colleagues 87 described a case where the cement fragment in inferior vena cava migrated to the right ventricle after 3 months of conservative treatment and led to fatal cardiac tamponade despite surgical repair. Plateker et al 78 documented transdiaphragmatic cement migration into the peritoneal cavity, which was extracted via a subxiphoid approach. Molloy T and colleagues 19 removed the embolized material by a robotic-assisted endoscopic approach. A 72-year-old woman 50 died within 3 h of symptom onset, and a 65-year-old man 59 refused treatment and was lost to follow-up.
Dead Patients’ Characteristics, Location of Embolus, Method of the Treatment and the Cause of Death
Discussion
The application of bone cement was gradually increasing with the widespread recognition of the clinical efficacy of VA, but the series of complications caused by bone cement cannot be ignored. As a rare but dreadful complication, ICE has been reported only in case report or case series report. Thus, the CARE guidelines and the JBI critical appraisal checklist were used to evaluate the quality of these reports. The CARE guidelines mainly aim to standardize the format and content of case reports, but magazines had their own requirements for these above. Some Items about report format, like Sub-item 1 (Title), Sub-item 3a (Introduction), Sub-item 3c (Conclusion) and Sub-item 13 (Informed consent) were evaluated as “Not reported + Not applicable” in more than 50% of studies. On the other hand, most items about patient information and diagnosis and treatment process were reported in more than 50% of studies, which ensured the implementation of our program. It was worth noting that this evaluation had partly revealed heterogeneity and bias. First, physical examination was not placed on a vital position (Sub-item 6). Second, the prognostic characteristics not being fully discussed, which has led to a reduced reference value (Sub-item 8d). Third, in most reports, follow-up and outcomes were defined as whether the patient was alive or not, lacking more detailed descriptions and important follow-up test result (Sub-item 10b and 10c). Fortunately, Sub-Item 10d (adverse and unanticipated events) was described in 26% of all the reports, which hinted about the uneventful follow-up and patient’s fully recovering. The JBI critical appraisal checklist revealed similar issues, and the rate of “Not reported + Not applicable” item for each case report or case series report was not exceeding 37.5%, which implied that the heterogeneity in content was not significant (Supplementary data, Table S3). Hence, strictly following the contents of the CARE guidelines (the reporting guidelines) and the JBI critical appraisal checklist (the guidelines for quality assessment) will be great help in decreasing heterogeneity and bias of case report (case series report).
The true incidence of ICE remains unclear primarily due to the absence of routine chest imaging and echocardiography after VA. Fadili Hassani S and colleagues 6 reported an ICE incidence after PVP was 3.9%, with symptomatic ICE occurring in 0.32%. However, this figure was likely underestimated for 2 reason. First, as a retrospective study, 16.4% of patients did not undergo chest imaging and were excluded from the analysis. Second, only 26.4% of patients had follow-up longer than 1 month so that the late-onset ICE may be mistaken as asymptomatic. The studies by Bohl MA 7 and Barakat AS9 indirectly reflected the incidence of symptomatic ICE, which were about 2.94% and 0.23% respectively. This discrepancy may be attributable to differences in surgical technique and the number of treated vertebrae. These findings suggested that the true ICE incidence may be higher than previously recognized and warranted further investigation. Postoperative monitoring for ICE should be considered in all patients undergoing VA.
Based on clinical appearance, ICE can be classified as symptomatic or asymptomatic. Symptomatic ICE manifests with clinical features such as dyspnea, chest pain, or arrhythmia, while asymptomatic ICE is more difficult to recognize. However, these manifestations lack specific and cannot distinguish ICE from PCE, cement toxicity, or the aggravation of preexisting cardiopulmonary diseases. Sometimes, these symptoms can be temporarily relieved with simple analgesia and oxygen inhalation, which may confuse clinicians and led to delay diagnosis.20,88 Symptomatic ICE may be caused by the following reasons. Firstly, mechanical injury and obstruction. The cement embolus will destroy the cardiac valve, perforate the heart wall, and cause cardiac-valve regurgitation and hemopericardium. Large cement emboli can also obstruct intracardiac blood flow directly. Secondly, co-presenting with PCE. Our review identified near 60 symptomatic cases with the presence of pulmonary embolism simultaneously. Thirdly, cytotoxicity. Anaphylactic toxins of cement could stimulate the cardiovascular system directly.
In our review, over 40 symptomatic patients experienced symptoms intraoperatively or within 48 hours postoperatively, while others experienced delayed symptom onset-ranging from several months up to 5 years. This suggests that ICE may follow a progressive pathological process in some patients, taking a long time for the clinical manifestation to become apparent. Thus, it cannot be determined whether the asymptomatic patients will remain asymptomatic for life or are on their way to be symptomatic. The lack of long-term follow-up data among the majority of asymptomatic patients in this study fed this fear.
The serum levels of inflammatory markers, cardiac enzymes and D-dimer are often normal limits or slightly elevated, but lack specificity. Marked elevations typically indicate the presence of other diseases. In our review, all patients were diagnosed by chest imaging with or without echocardiography, except for 1 identified through autopsy
50
and another diagnosed by three-dimensional echocardiography only.
76
Beyond question, echocardiography with contrast-enhanced CT was an efficient diagnostic approach, but its radiation precluded routine screening. Thus, we recommend conducting postoperative chest X-ray radiography and comparing them with the corresponding preoperative images in order to aid in the screening. If a patient developed chest discomfort during or after VA, clinicians must be vigilant and ensure that the patient underwent CT as early as possible (Figure 2). The simplified decision-making flow chart for the management of intracardiac cement embolism * The signs of bone cement leakage referred to any situation where bone cement leaked outside the vertebral body. ** Not recommended
Fadili Hassani S et al 6 identified a higher number of treated vertebrae during a single PVP procedure as an independent risk factor for ICE. Additionally, high-quality fluoroscopic guidance was associated with a lower ICE rate. However, factors such as age, gender, vertebral lesion type and location and operator’s experience did not influence ICE risk. To our knowledge, this study is the only 1 to investigate the risk factors of ICE following VA. Learning from the experience of PCE which had the same cement migration pathway as ICE, risk factors for PCE included PVP, thoracic vertebra, higher cement volume injected per level, more than 3 vertebrae treated per procedure, and venous cement leakage.3,108 Considering the migration trajectory, it was not surprising that most studies had established a correlation between venous cement leakage and PCE. Guo H et al 108 further figured that venous leakage can be recognized as an early sign of PCE and a signal to stopping the injection. In our review, nearly 50 cases clearly documented venous leakage, and in 5 special cases, bone cement had hardened in the paravertebral vein or inferior vena cava, fractured, and then launched itself like a missile down the blood vessels and toward the heart at some point.35,78,86,87,89 1 option under consideration was the prophylactic deployment of a retrievable inferior vena cava filter to prevent the migration of paravertebral vein cement or inferior vena cava cement. 109 Based on the above risk factors, recommendations include injecting cement with a toothpaste-like viscosity in small doses multiple times under continuous fluoroscopic monitoring, reducing cement volume, limiting the number of augmented vertebral body, and positioning the tip of screw/puncture needle away from the vertebral midline to reduce paravertebral venous leakage.
Treatment should be individualized, but most reports lacked objective evidence for making the treatment plan (Sub-Item 8d of CARE guidelines). Generally, asymptomatic patients required close monitoring without intervention. The role of anticoagulation remained controversial.40,54,67 Some scholars advocated that anticoagulation helped cement endothelialization and stopped being thrombogenic, while others thought that this theory lacked experimental evidence and anticoagulation increased the risk of bleeding complications. In cases of cardiac perforation, anticoagulation may worsen outcomes by promoting haemopericardium. In our series, anticoagulation conferred no additional benefit. Therefore, its use must be carefully weighed against potential risks.
For symptomatic patients, surgical embolectomy is a viable option. Indications included cardiac chamber perforation, impending perforation, haemorrhage, and embolization-induced arrhythmia.19,44,66 In our reviewed cases, open-heart surgery was the most frequently chosen for symptomatic patients and was reported to be successful. Only 2 cases died postoperatively from hepatic failure 82 and multiple organ failure 83 despite successful embolectomy. One patient experienced intraoperative cement fragmentation with subsequent asymptomatic pulmonary vascular migration. 81 The first endovascular extraction of cardiac cement embolism under fluoroscopic observation was reported by Braiteh F and Row M in 2009. 84 Since then, 16 similar cases had been reported, but only 5 patients9,10,47,84,85 completed successfully without any accidents. Hatzantonis C and colleagues 48 hold that percutaneous retrieval was more suitable for symptomatic patients with right atrium thrombi. Due to unpredictable further cement embolism fragmentation and distal embolization, and, importantly, inability to repair the cardiac puncture site, percutaneous retrieval did not replace open-heart surgery. Park JS and colleagues 51 proposed a hybrid approach, in which the endovascular procedure captured the cement and moved it as distally as possible from the heart, followed by open-vessel surgery to extract the cement. This strategy may offer a safer alternative for high-risk patient to minimize the damage of open-heart surgery and shorten recovery time.
To date, no consensus exists on the management of ICE so that we developed a simplified decision-making flowchart (Figure 2) based on the reviewed literature. In this flowchart, patients undergoing VA were first categorized into 4 sub-types according to the signs of cement leakage and cardiopulmonary symptoms. Cement leakage referred to any extrusion of cement beyond the vertebral body, including leakage into the endplate, paravertebral space, or epidural/paravertebral veins. Subsequent finding on CT and echocardiography then guide the selection of treatment plan. This flowchart was hoped to assist clinical decision-making in managing ICE.
Limitation
Our review had several limitations. First, no randomized controlled trial was included so that meta-analysis can not be assessed and some requirement from the PRISMA 2020 checklist were not met. This reflects not only the rarity of ICE, but also the need for greater concentration. Second, risk factors for ICE were not fully explored. Consequently, lessons from PCE should be applied to help reduce ICE incidence. Finally, the diagnosis, treatment and follow-up of ICE had not been standardized, introducing uncontrollable information bias. Especially, most reports defined follow-up and outcomes solely by patient survival, without detailed descriptions. Therefore, further high-quality prospective cohort study is warranted to offer more convincing conclusions. Despite these limitations, our findings still offer useful insights for clinical practice.
Conclusion
ICE was a rare but dreadful complication of VA. Its clinical presentation and blood-related biochemical parameters were nonspecific. Most patients were diagnosed by chest imaging. Comparative analysis of pre- and postoperative chest X-ray radiography was recommended as a screening method. As no clear risk factors for ICE have been established, clinical experience with PCE may help inform preventative strategies. Based on current literature, we developed a simplified decision-making flowchart to guide ICE management (Figure 2). In general, asymptomatic patients required close monitoring without intervention, whereas symptomatic patients needed surgical embolectomy. We believe our study may provide some valuable insights into the clinical practice of this complication.
Supplemental Material
Supplemental Material - Incidence, Clinical Characteristics, Treatment and Outcomes of Intracardiac Cement Embolism After Vertebral Augmentation: A Systematic Review
Incidence, Clinical Characteristics, Treatment and Outcomes of Intracardiac Cement Embolism After Vertebral Augmentation: A Systematic Review by Hao Tong, MD, Peng Xie, MD, Tao Song, MD, Haocheng Ma, MD, Quanzhou Xiao, MD, Zhehao Dai in Global Spine Journal
Footnotes
Ethical Approval
Ethical approval is not required for this study.
Authors’ Contributions
All authors reviewed and approved the final manuscript. Hao Tong and Peng Xie have contributed to this paper equally. Zhehao Dai: Conceptualization, Funding acquisition, Project administration, Supervision, Writing–review&editing. Hao Tong: Methodology, Writing–original draft. Peng Xie: Formal analysis, Writing–original draft. Tao Song: Data curation, Investigation. Haocheng Ma: Data curation, Formal analysis, Investigation. Quanzhou Xiao: Data collection and shift, Data check.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Graduate Education and Teaching Reform Project of Central South University (2024JGB087), Soft Science Research Program Project of Changsha (kh2502020).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data underlying this article are available from the corresponding author upon reasonable request within 6 months after publication.
Declaration of Generative AI and AI-assisted technologies in the Writing Process
During the preparation of this work, the authors did not use generative artificial intelligence (AI) and AI-assisted technologies.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
