nueva página del texto (beta)
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Introduction
The endovascular balloon occlusion is a method that has been introduced as a concept for more than 70 years, a result of the Korean War,1 and since then, it has been put into sequential study and applied in the context of patient victims of military traumatic injuries and civilians. This has advanced to the point of becoming an endovascular technique that is easily applicable in the emergency department and in surgery rooms.2 However, in search of making better use of this resource, in recent years, the concept of intravascular occlusion in the arterial and/or venous system has been applied in elective surgery as a complement in the treatment of various medical-surgical conditions in search of explored new strategies to prevent and control intra- and postoperative hemorrhagic events.
Historically, one of the main and most feared complications related to fatal outcomes is massive bleeding and secondary hypovolemic shock. Numerous efforts have been directed towards creating devices and interventional techniques to reduce or even prevent bleeding. Since its first application by Hughes on two wounded soldiers in the Korean War,1 large blood vessel endovascular occlusion devices have opened multiple doors, leading to two main objectives. The first is bleeding control, such as in cases of severe trauma associated with non-compressible bleeding, ruptured abdominal aortic aneurysms, postpartum hemorrhage, etc., where the goal is to control an already established hemorrhage.3 The second is the prevention of bleeding, which is the focus of more recent research, implementing these devices in elective (non-urgent) pelvic-obstetric, renal, hepatobiliary, and gastrointestinal procedures to avoid hemorrhage and its associated complications.
Effective management of intraoperative bleeding is essential for the success of any surgical procedure and remains a significant concern worldwide and locally, as hemorrhagic complications continue to impact the outcomes of certain types of procedures negatively, increasing reinterventions, massive transfusions, organ or multi-organ dysfunction, and in some situations, even death.4
In recent decades, technological advancements have led to the development of innovative endovascular techniques and devices to control hemorrhage during surgery. One of these significant advancements has been the introduction and use of intravascular occlusion balloons in adult patients undergoing elective surgery.3),(4 Initially, these devices were used in the trauma context to control massive bleeding, either through open or closed methods, but their application has expanded to non-traumatized adult patients undergoing major elective surgical procedures with significant bleeding risk.5
The implementation of these occlusion devices at the aortic and vena cava levels has demonstrated substantial benefits in terms of improved survival rates and fewer post-surgical hemorrhagic complications in patients who have experienced open or closed accidents.6,7 This initial success has led to the exploration of their preventive use in elective surgeries, where a high risk of potentially fatal intraoperative bleeding is anticipated.8
Despite the growing popularity of this technique and several isolated studies, there is little evidence in the scientific literature regarding its indications, efficacy, safety, impact on the magnitude of bleeding, blood component polytransfusion, and mortality in the context of elective surgeries. Moreover, its current use is justified by isolated studies with diverse methodologies and results, often based on local experiences. Therefore, it is crucial to address this knowledge gap in an organized and systematic manner so that the available data can be collected, analyzed, and interpreted, and based on this, establish guidelines based on the best evidence to optimize bleeding outcomes for patients undergoing elective surgical procedures.
This systematic literature review aims to provide a specific response by thoroughly evaluating existing studies on the use of intravascular occlusion balloons in elective surgeries concerning the amount of intraoperative bleeding. By doing so, it seeks to provide a more synthesized and concrete view of the actual effectiveness of this technique in the context of non-traumatic surgical procedures based on the literature available to date.
The results of this review could have significant implications for clinical practice. If intravascular occlusion balloons are confirmed to be effective in reducing intraoperative bleeding in elective surgeries, this could support their prophylactic use and lead to a substantial reduction in bleeding-related complications. Additionally, by providing evidence-based guidance, this review can serve as a foundation for developing local, national, and international clinical protocols and help scientific communities generate recommendations on the implementation of this technique.
Material and methods
Methodology
This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist.9
The inclusion criteria for the review encompassed studies involving adult patients over 18 years of age undergoing elective surgery where intravascular occlusion balloons were used. The types of interventions considered included intravascular occlusion with arterial or venous balloons in elective surgery. The primary outcome of interest was the amount of intraoperative hemorrhage when intravascular balloon occlusion was utilized. Eligible studies included observational, analytical, and descriptive types and only those reported in English and Spanish were considered.
The exclusion criteria for the review included studies carried out in animal models or species other than humans, as well as those focused on emergency surgery. Studies were also excluded if they used devices other than intravascular, temporary, or definitive occlusion balloons in elective surgery or if they presented unrelated results that did not provide relevant and clear information on the reduction of intraoperative hemorrhage with intravascular occlusion balloons. Duplicate studies were excluded, retaining only the most complete and detailed version. Additionally, publications not subject to review by both researchers were excluded.
The search was carried out in three main databases: PubMed, Ovid, Embase, and Google Scholar, which is a gray literature database, and the National Institute of Health (INS) as a national database. Articles written in English and/or Spanish were accepted and published in the last 10 years until April 2024. The search result was stored in Mendeley and Rayyan© as organizer and reference manager, respectively. Additionally, bibliographic references of the included studies were searched and compiled to ensure a comprehensive review of the literature.
The searches were executed with the keywords in terms Mesh (Medical Subject Headings) Thesaurus on the health sciences of the National Library of Medicine (NLM); Vena Cava, Superior, Vena Cava, Inferior, Aorta, Aorta, Thoracic, Aorta, Abdominal, Balloon Occlusion, Wounds, and Injuries. Once these Mesh terms were set, the Boolean operators were used as follows: (((Vena Cava, Superior[Mesh] OR Vena Cava, Inferior[Mesh]) OR (Aorta”[Mesh] OR Aorta, Thoracic[Mesh] OR Aorta, Abdominal[Mesh])) AND Balloon Occlusion[Mesh]) NOT Wounds and Injuries[Mesh]. Duplicate studies were removed using the Rayyan tool.
Study selection
Each author independently reviewed the titles and abstracts of the articles in the database obtained as a result of the search strategy; Articles that were not related to the research question were excluded. Full texts were obtained only from articles considered potentially eligible by at least one reviewer. Subsequently, each author independently reviewed the full texts of the potentially eligible articles, verified the inclusion and exclusion criteria, and established the definitive articles for carrying out the present systematic review. Cases where there was a discrepancy were resolved by consensus in the first instance, and if disagreement persisted, a third reviewer determined whether or not to include the article.
Data extraction process
For data extraction, the artificial intelligence tool SciSpace10 was used as the first instance, where the articles included in the study were entered and the specific data were screened. The information collected from each article included authors, year of publication, study design, number of participants in the intervention group and control group, indication for intravascular occlusion, site of vascular occlusion, amount of bleeding during the intervention, surgical time, and complications associated with the intervention. Additionally, for some numerical variables, dispersion measures such as the mean and standard deviation were recorded. In cases where automatic extraction of information was not obtained, it was added manually. Subsequently, each author independently corroborated the veracity of the information collected.
The data obtained were tabulated in a standardized Excel spreadsheet (Microsoft). Finally, articles that did not provide the total of the mentioned variables were excluded in order to avoid bias in obtaining results and analyzing them.
Risk of bias and quality of included studies
The Newcastle-Ottawa scale was used to assess the risk of bias, a validated and widely used instrument to evaluate the risk of bias in observational studies.11 This scale considers three domains: selection of participants, comparability between groups, and evaluation of exposure or results.12 Methodological quality was classified according to the following criteria: (a) Good: three to four stars in selection, one to two in comparability, and two to three in results/exposition; (b) Fair: two stars in selection, one to two in comparability and two to three in results/exhibition; (c): zero to one star for selection, zero for comparability and zero to one for results/exhibition.11
Statistical analysis
Data analysis and management were carried out using STATA statistical software. For the qualitative variables, the log Odds ratio method was used, with a random effects model to calculate the Odds Ratio (OR) and the 95% confidence interval (95% CI). For numerical variables, the inverse variance method with a random effects model was used to determine the standardized mean difference (SMD) with its 95% CI. The presence of statistical heterogeneity was evaluated using the I2 test to measure the magnitude of heterogeneity; statistical heterogeneity was considered a value greater than 50%.
Results
After applying the search strategy across all databases, we obtained a total of 2,592 articles, distributed as follows: PubMed (n = 403), EMBASE (n = 845), OVID (n = 585), Google Scholar (n = 758), and INS (n = 0). Ultimately, we included 17 articles in the systematic review, excluding the remaining publications through a rigorous screening process (Figure 1).
Our analysis revealed no randomized controlled trials regarding occlusion balloons in elective surgery. Among the 17 included studies, 10 were retrospective cohorts,13-22 one was a prospective cohort study,23 and six were case-control studies.24-29 We assessed the quality and risk of bias of all studies using the Newcastle-Ottawa Scale for cohorts and case-control studies, focusing on selection, comparability, and outcomes. Two retrospective cohorts exhibited a moderate risk of bias, while the remaining articles demonstrated good quality and low risk of bias (Table 1).
Table 1: Results of quality assessment using the Newcastle-Ottawa scale for all studies.
| Study | Type of study | Selection | Comparability | Exposure or results | Methodological quality |
|---|---|---|---|---|---|
| Ioscovich A (2023) | Cases and controls | 4 | 2 | 3 | Good |
| Hao Z (2016) | Cases and controls | 4 | 1 | 3 | Good |
| Zeng C (2017) | Cases and controls | 3 | 1 | 1 | Regular |
| Filho S (2019) | Retrospective cohort | 4 | 1 | 3 | Good |
| Huo F (2021) | Retrospective cohort | 4 | 1 | 2 | Good |
| Kaneda H (2017) | Cases and controls | 4 | 1 | 2 | Good |
| Kyozuka H (2023) | Retrospective cohort | 4 | 1 | 3 | Good |
| Papillon-Smith J (2020) | Retrospective cohort | 4 | 1 | 3 | Good |
| Ye Y (2023) | Retrospective cohort | 4 | 1 | 3 | Good |
| Wu Q (2016) | Retrospective cohort | 4 | 1 | 3 | Good |
| Peng W (2020) | Retrospective cohort | 2 | 1 | 2 | Regular |
| Zhao X (2016) | Cases and controls | 4 | 2 | 2 | Good |
| Duan X (2018) | Retrospective cohort | 3 | 1 | 2 | Good |
| Wang Y (2020) | Retrospective cohort | 4 | 1 | 3 | Good |
| Peng Y (2020) | Cases and controls | 4 | 2 | 3 | Good |
| Zhao Z (2020) | Prospective cohort | 3 | 1 | 2 | Good |
| Zangh Y (2018) | Retrospective cohort | 4 | 1 | 3 | Good |
Good: 3 to 4 stars for selection, 1 to 2 for comparability, and 2 to 3 for results/exhibition; Fair: 2 stars in selection, 1 to 2 in comparability and 2 to 3 in results/exhibition; Bad: 0 to 1 on selection, 0 on comparability and 0 to 1 on results/exposure.
This review included a total of 3,379 patients, comprising 157 men (4.7%) and 3,222 women (95.3%). The average age in the intervention group was 36.4 years (SD: 5.9), compared to 37.0 years (SD: 6.1) in the non-intervention group (Table 2).
Table 2: Comparative outcomes in groups studied with respect to age, indication and site of occlusion.
| Study | Patients | Intervention group (ball) | Non-intervention group (no ball) | Indication | Occlusion site | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| n | Gender | Age (years) Mean ± SD |
n | Gender | Age (years) Mean ± SD |
||||||
| H | M | H | M | ||||||||
| Ioscovich A 2023 | 21 | 10 | 0 | 10 | 35 ± 5.02 | 11 | 0 | 11 | 33.8 ± 4.5 | Abnormal placentation | Infrarenal abdominal aorta |
| Hao Z 2016 | 41 | 18 | 11 | 7 | 34.2 ± 2.5 | 23 | 14 | 9 | 34 ± 2.1 | Complex acetabular fracture | Infrarenal abdominal aorta |
| Zeng C 2017 | 86 | 48 | 0 | 48 | 32.3 ± 5.27 | 38 | 0 | 38 | 33.1 ± 5.23 | Abnormal placentation | Infrarenal abdominal aorta |
| Filho S 2019 | 35 | 28 | 0 | 28 | 33 (24-43)* | 7 | - | - | - | Abnormal placentation | Bilateral internal iliac artery |
| Huo F 2021 | 33 | 17 | 0 | 17 | 32.82 ± 4.45 | 16 | 0 | 16 | 34.44 ± 4.79 | Abnormal placentation | Infrarenal abdominal aorta |
| Kaneda H 2017 | 518 (Hysterectomy) | 12 | 0 | 12 | 49.5 (36-62)* | 506 | 0 | 506 | 47 (34-69)* | Large uterine cervical fibroid | Bilateral internal iliac artery |
| 305 (Miomectomy) | 10 | 0 | 10 | 35.5 (28-40)* | 295 | 0 | 295 | 38 (23-63)* | |||
| Kyozuka H 2023 | 37 | 13 | 0 | 13 | 37.5 (30.8-41)* | 24 | 0 | 24 | 35 (32-38)* | Abnormal placentation | Supraceliac Aorta |
| Papillon-Smith J 2020 | 79 | 47 | 0 | 47 | 35 (22-51)* | 32 | 0 | 32 | 34 (25-44*) | Abnormal placentation | Internal iliac artery |
| Ye Y 2023 | 364 | 278 | 0 | 278 | 34 (30-37)* | 86 | 0 | 86 | 34 (32-36)* | Abnormal placentation | Infrarenal abdominal aorta |
| Wu Q 2016 | 268 | 230 | 0 | 230 | 29.5 ± 3.6 | 38 | 0 | 38 | 30.4 ± 4 | Abnormal placentation | Infrarenal abdominal aorta |
| Peng W 2020 | 586 | 252 | 0 | 252 | 32.69 ± 4.62 | 296 | 0 | 296 | 32.74 ± 4.84 | Abnormal placentation | Infrarenal abdominal aorta |
| Zhao X 2016 | 57 | 23 | 13 | 10 | 44.36 ± 13.34 | 34 | 15 | 19 | 45.41 ± 15.7 | Pelvic or hip tumor resection | Infrarenal abdominal aorta |
| Duan X 2018 | 45 | 22 | 0 | 22 | 32,1 ± 6.9 | 23 | 0 | 23 | 31.7 ± 8.5 | Abnormal placentation | Infrarenal abdominal aorta |
| Wang Y 2020 | 623 | 623 | 0 | 623 | - | 23 | 0 | - | - | Abnormal placentation | Infrarenal abdominal aorta |
| Peng Y 2020 | 104 | 48 | 0 | 48 | 32.08 ± 3.94 | 56 | 0 | 48 | 33.46 ± 4.53 | Abnormal placentation | Bilateral internal iliac artery |
| Zhao Z 2020 | 121 | 57 | 33 | 24 | 48 (18-70)* | 64 | 34 | 30 | 45 (18-70)* | Pelvic or hip tumor resection | Infrarenal abdominal aorta |
| Zangh Y 2018 | 56 | 30 | 20 | 10 | 42 ± 18 | 26 | 17 | 9 | 50 ± 19 | Pelvic or hip tumor resection | Infrarenal abdominal aorta |
* Median and (range)
SD = Standard deviation.
The conditions for which intravascular occlusion balloons were utilized in elective surgical treatments included abnormal placentation (placenta accreta) in 2,281 patients (67.5%), giant cervical uterine fibroids in 823 patients (24.3%), resection of sacrococcygeal tumors in 234 patients (6.9%), and open reduction and internal fixation of complex acetabular fractures in 41 patients (1.2%) (Table 2).
The anatomical sites selected for endovascular balloon occlusion were infrarenal abdominal aorta in 68% (n = 2,301), bilateral internal iliac arteries in 30.8% (n = 1,041), and suprarenal aorta in 1.09% (n = 37).
Regarding intraoperative bleeding, the intervention group (endovascular occlusion balloon use) had an average blood loss of 1,256 mL (SD: 669.9), while the non-intervention group (no balloon use) reported an average blood loss of 2,112 mL (SD: 1,027.8) (Table 3).
Table 3: Comparative outcomes in groups studied with respect to amount of bleeding.
| Study | Intervention group (ball) | Non-intervention group (no ball) | ||||||
|---|---|---|---|---|---|---|---|---|
| Bleeding (mL) | Time (min) | Bleeding (mL) | Time (min) | |||||
| Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
| Ioscovich A 2023 | 1,060 | 296.64 | 119 | 29.41 | 4,400 | 2,787 | 149.81 | 47.69 |
| Hao Z 2016 | 1,247.2 | 67.1 | 213.3 | 8.9 | 1,526.1 | 69.9 | 248.30 | 7.00 |
| Zeng C 2017 | 1,467.71 | 1,075.77 | 92.19 | 32.5 | 2,218.42 | 1,572.2 | 119.47 | 59.37 |
| Filho S 2019 | 1,193 | 679 | 332 | 70 | 2,273.4 | – | – | – |
| Huo F 2021 | 3,167.65 | 3,255.71 | – | – | 2,831.25 | 1,906.03 | – | – |
| Kaneda H 2017 | 510 | – | 178 | 116-300* | 350 | – | 165.50 | 57-686 |
| 727.5 | – | 157.5 | 156-218* | 390 | – | 160 | 52-366* | |
| Kyozuka H 2023 | 1,110 | – | 144 | 112-163* | 2,160 | – | 146 | 126-164* |
| Papillon-Smith J 2020 | 1,713 | 181 | 353 | 14.00 | 1,874 | 245 | 227 | 13.00 |
| Ye Y 2023 | 1,370.5 | 752 | 96.3 | 37.6 | 3,536.8 | 1,383.2 | 160.60 | 45.50 |
| Wu Q 2016 | 921 | 199 | 64.1 | 5.1 | 2,790 | 335 | 92.10 | 9.70 |
| Peng W 2020 | 1,967.66 | 1,466.64 | 191.05 | 59.4 | 1,338.18 | 1,286.14 | 153.02 | 57.33 |
| Zhao X 2016 | 437.23 | 54.32 | 193.28 | 63.47 | 1,846.45 | 87.56 | 273.63 | 73.31 |
| Duan X 2018 | 597 | 359 | 63.8 | 12.3 | 2,687 | 575 | 118.80 | 22.40 |
| Wang Y 2020 | 620 | 570 | 65.3 | 14.5 | 2,687 | 575 | – | – |
| Peng Y 2020 | 1,504.17 | 1,123.44 | 158.44 | 57.32 | 1,108.04 | 1,008.32 | 104.20 | 46.22 |
| Zhao Z 2020 | 1,000 | – | 185 | 100-500* | 1,350 | – | 260 | 180-600* |
| Zangh Y 2018 | 2,000 | – | 215 | 110-430* | 2,650 | – | 225 | 115-340* |
* Range.
SD = Standard deviation.
Surgical duration also constituted a measured outcome in this study. In the intervention group, the average duration of surgical procedures was 165.9 minutes (SD: 84), while in the non-intervention group, it was 173.5 minutes (SD: 58.6).
A total of 149 participants (4.4% of the total included) experienced complications related to the use of the endovascular occlusion balloon. These included 79 arterial thromboembolic events, 45 emergency hysterectomies due to uncontrolled bleeding, 12 cases of skin and subcutaneous tissue injuries or local hematomas, five cases of vasospasm, two pseudoaneurysms of femoral vessels, and two cases related to balloon issues (migration and rupture). Other less frequent complications included femoral nerve injury, arteriovenous fistula, surgical site infection, and femoral artery dissection (one case each). Five studies reported no complications (Table 4).
Table 4: Intervention group with respect to complications.
| Complication | n (%) |
|---|---|
| Arterial thromboembolism | 79 (53.00) |
| Emergency hysterectomy | 45 (30.20) |
| Skin lesions or local hematoma | 12 (8.05) |
| Vasospasm | 5 (3.35) |
| Femoral pseudoaneurysm | 2 (1.34) |
| Balloon dysfunction | 2 (1.34) |
| Femoral nerve injury | 1 (0.67) |
| Arteriovenous fistula | 1 (0.67) |
| Operative site infection | 1 (0.67) |
| Arterial dissection | 1 (0.67) |
| Total | 149 (100.00) |
Quantitative Analysis
The systematic review (meta-analysis) was performed on 17 studies encompassing a total of 3,379 patients. The following variables were analyzed in the meta-analysis.
Intraoperative Bleeding
In this outcome, only 12 studies were subjected to statistical analysis (Table 5), revealing that the intervention serves as a protective factor against bleeding, with an odds ratio (OR) of −3.43 (95% CI −6.22; −0.63). The overall analysis exhibited high statistical heterogeneity (I2 of 99.7%) with a significant p-value (Figure 2).
Table 5: GRADE level of evidence and certainty. Intravascular balloon occlusion compared with not using intravascular occlusion balloon to reduce intraoperative bleeding.
| Certainty assessment | N of patients | Effect | Certainty | Importance | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N of studies | Study design | Risk of bias | Inconsistency | Indirectness | Imprecision | Other considerations | intravascular occlusion balloon | not intravascular occlusion balloon | Relative (95% CI) |
Absolute (95% CI) |
||
| Ioscovich A 2023 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Strong association | 10 cases 11 controls |
OR −1.58 (−2.53 to −0.63) |
− | ⨁⨁⨁◯ Moderate |
Important | |
| − | 0.0% |
0 fewer per 1,000 (from 0 fewer to 0 fewer) |
||||||||||
| Hao Z 2016 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Very strong association | 18 cases 23 controls |
OR −3.98 (−5.03 to −2.93) |
− | ⨁⨁⨁⨁ High |
Important | |
| − | 0.0% |
0 fewer per 1,000 (from 0 fewer to 0 fewer) |
||||||||||
| Zeng C 2017 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Very strong association | 48 cases 38 controls |
OR −0.56 (−0.99 to −0.13) |
− | ⨁⨁⨁⨁ High |
Important | |
| − | 0.0% |
0 fewer per 1,000 (from 0 fewer to 0 fewer) |
||||||||||
| Huo F 2021 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Very strong association | 17/33 (51.5%) | 16/33 (48.5%) |
OR 0.12 (−0.54 to 0.79) |
383 fewer per 1,000 (from 1,000 fewer to 58 fewer) |
⨁⨁⨁⨁ High |
Important |
| Papillon−Smith J 2020 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Strong association | 47/79 (59.5%) | 32/79 (40.5%) |
OR −0.76 (−1.22 to −0.30) |
1,000 fewer per 1,000 (from 1,000 fewer to 662 fewer) |
⨁⨁⨁◯ Moderate |
No important |
| Ye Y 2023 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Strong association | 278/364 (76.4%) | 86/364 (23.6%) |
OR −2.30 (−2.60 to −2.01) |
1,000 fewer per 1,000 (from 1,000 fewer to 1,000 fewer) |
⨁⨁⨁◯ Moderate |
Important |
| Wu Q 2016 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Very strong association | 230/268 (85.8%) | 38/268 (14.2%) |
OR −8.36 (−9.15 to −7.57) |
1,000 more per 1,000 (from 1,000 more to 1,000 more) |
⨁⨁⨁⨁ High |
Important |
| Peng W 2020 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Strong association | 252/586 (43.0%) | 296/586 (50.5%) |
OR 0.46 (0.29 to 0.63) |
186 fewer per 1,000 (from 277 fewer to 114 fewer) |
⨁⨁⨁◯ Moderate |
Important |
| Zhao X 2016 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Strong association | 23 cases 34 controls |
OR −18.28 (−21.68 to −14.89) |
− | ⨁⨁⨁◯ Moderate |
No important | |
| − | 0.0% |
0 fewer per 1,000 (from 0 fewer to 0 fewer) |
||||||||||
| Duan X 2018 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Strong association | 22/45 (48.9%) | 23/45 (51.1%) |
OR −4.26 (−5.31 to −3.21) |
778 more per 1,000 (from 709 more to 913 more) |
⨁⨁⨁◯ Moderate |
Important |
| Wang Y 2020 | Non-randomised studies | Not serious | Not serious | Not serious | Serious | Strong association | 600/623 (96.3%) | 23/623 (3.7%) |
OR −3.62 (−4.08 to −3.16) |
198 fewer per 1,000 (from 222 fewer to 175 fewer) |
⨁⨁◯◯ Low |
No important |
| Peng Y 2020 | Non-randomised studies | Not serious | Not serious | Not serious | Not serious | Strong association | 48/104 (46.2%) | 56/104 (53.8%) |
OR 0.37 (−0.02 to 0.76) |
237 fewer per 1,000 (from 562 fewer to 68 fewer) |
⨁⨁⨁◯ Moderate |
Important |
The general asymmetry of the funnel plot for this variable suggests significant publication bias. However, the dispersion observed in smaller studies may indicate heterogeneity among them (Figure 3).
The lack of alignment of most studies along the regression line suggests general disparity in the meta-analysis results, characterized by marked heterogeneity (Figure 4).
Gender
No significant differences were found, with an OR of 0.19 (95% CI −0.26; 0.64), and no statistical heterogeneity was observed (I2 of 0%) (Figure 5).
Age
No significant differences were identified with respect to age, with an OR of 0.06 (95% CI −0.15; 0.04), and no statistical heterogeneity was observed (I2 of 0%) (Figure 6). The general symmetry of the funnel plot suggests an absence of relevant publication bias. However, the slight dispersion observed in smaller studies may indicate some heterogeneity, warranting further exploration to identify potential differences in study designs or populations (Figure 7). The alignment of most studies along the regression line indicates overall consistency in the meta-analysis results, with limited heterogeneity and no outlier studies contributing to the overall heterogeneity (Figure 8).
Surgical duration
No significant differences were found in terms of duration reduction, with an OR of −0.47 (95% CI −2.13; 1.18), and statistical heterogeneity was observed (I2 of 99.5%) (Figure 9). The overall asymmetry of the funnel plot for this variable suggests significant publication bias. However, the dispersion in smaller studies may indicate heterogeneity among them (Figure 10). The lack of alignment of most studies along the regression line suggests general disparity in the meta-analysis results, characterized by marked heterogeneity (Figure 11).
Discussion
Our results indicate that certain sociodemographic variables, such as female gender and age, characterize the population in which the intravascular balloon occlusion technique is most frequently applied. These findings align with the observational study by Wang Y,21 which included approximately 623 female patients, and the study by Peng W19 with 296 female patients. Both studies were conducted by gynecology groups focusing on abnormal placentation pathologies. Age, as an isolated variable, is supported by studies from Ye Y,17 Wu Q,18 Peng W,19 and Wang Y,21 which suggest that abnormal placentation is more prevalent among young women of reproductive age, typically under 40 years.
The pathology most frequently addressed using this technique was abnormal placentation, corroborated by studies such as those by Peng W19 and Ye Y.17 This was followed by the presence of giant fibroids, as highlighted in Kaneda H’s study,27 which reported the largest patient cohorts and emphasized the technique’s role in reducing morbidity and mortality. Notably, the third most common pathology was non-gynecological in nature, specifically the open reduction of long bone fractures in the lower extremities, as reported by Hao Z.25 This study also included the highest number of male patients in our review.
Regarding the occlusion sites, all procedures were performed within the arterial system, primarily at the infrarenal aorta. In two studies, Kaneda H27 and Peng W,19 occlusions were performed as distally as possible, bilaterally at the internal iliac arteries. These findings correspond with studies that had the largest patient populations and most representative pathologies, specifically abnormal placentation and gynecological-pelvic tumors.
Concerning bleeding, the primary variable studied, our findings indicate that the application of prophylactic occlusion effectively reduced intraoperative bleeding. This result aligns with studies such as those by Zhao X,28 Duan X,20 Zeng C,26 and Ye Y,17 demonstrating that this method helps control and prevent intraoperative hemorrhage, thereby improving outcomes.
The most common complication observed was transient arterial thrombotic events, which were resolved with medical management. This was followed by emergency conversions to hysterectomy due to uncontrolled bleeding, predominantly in patients with abnormal placentation. However, this complication rate did not exceed 5% of the included population, suggesting that the risks associated with using this technique in elective surgery are acceptable.
Regarding the limitations of our research, we note the absence of randomized controlled trials specifically addressing the primary outcome of bleeding. Additionally, significant heterogeneity existed among the included studies, which we attempted to address through various stratification methods and statistical analyses.
In summary, our study suggests that the intravascular occlusion technique is effective in reducing intraoperative bleeding and may have significant clinical applications. Nevertheless, further research through controlled clinical trials is necessary to establish clear diagnostic inclusion criteria for participants and to individualize outcomes based on specific interventions within our population, thereby confirming these findings.
Conclusions
Intravascular balloon occlusion effectively reduces intraoperative blood loss in elective general non-cardiac surgical procedures with a high risk of hemorrhage. The most common complications include transient arterial thrombosis and localized issues at the puncture site.
While various clinical scenarios exist for applying this technique, the strongest evidence focuses on obstetric procedures, particularly those involving abnormal placentation.
Further research is essential, especially controlled clinical studies that establish clear diagnostic inclusion criteria for participants. Additionally, individualizing results based on specific interventions within our population is crucial to validate these findings.
