Effectiveness of REBOA in Reducing Blood Loss During Cesarean Section in Patients with PAS

Study design and study environment This will be a prospective, randomized controlled trial (RCT) that will investigate the superiority of REBOA in combination with cesarean section compared to the traditional cesarean section for pregnant women with PAD disorders. The study that will take place in the National Research Center for Mother and Children (NRCMC), University Medical Center (UMC), Astana, Kazakhstan. NRCMC is a tertiary care leading specialized institution in the field of obstetrics and gynecology in Kazakhstan. The hospital has sufficient experience in the field of scientific research, and a powerful technical and personnel base. Moreover, all UMC hospitals are part of Nazarbayev University Medical cluster (NU Medicine), which employs leading professors and researchers in the field of obstetrics and gynecology. The Center has a Department of Functional Diagnostics and Interventional Radiology with highly qualified specialists who have been trained at the international level. The NRCMC has all clinical units for patient admission and proper healthcare, 24/7 duty service of obstetricians-gynecologists and anesthesiologists-resuscitators. The diagnostic department has ultrasound machines with Doppler scanning opportunity (Philips, with a 9.4 Hz vaginal probe and a 5.1 Hz convex probe with Doppler) and magnetic resonance imaging machine (MRI), (PHILIPS Ingenia 3.0 T). The transfusiology department of NRCMC has a TEG5000 thromboelastograph device complete with a computer and a Haemonetics 2017 printer.

Study subjects The study population will consist of pregnant women who will be admitted to NRCMC of UMC for cesarean section due to placenta previa complicated by PAS disorders. Study subjects will be divided in cases and controls. The following inclusion criteria will be followed for case group: (1) singleton pregnancy at 34 weeks of gestation or more, (2) age 18 to 45 years, (3) confirmed diagnosis of placenta accreta spectrum, (4) indication for elective cesarean section, (5) consent for aortic balloon occlusion during cesarean section. Women who will not agree to undergo the aortic balloon occlusion procedure during cesarean section will be included in the control group.

The following exclusion criteria will be followed for both, case and control groups: (1) pregnant patients at the gestational age ˂ 34 weeks of gelation, (2) multiple pregnancy, (3) need for emergency caesarean section, (4) coagulopathy (known coagulopathy, women with abnormal baseline activated partial thromboplastin time (aPTT), prothrombin time (PT), international normalization ratio (INR) before surgery). In addition, patients on antiplatelet and anticoagulant drugs with known history of thrombotic events will be excluded.

Recruitment process The study participant recruitment will be performed from December 2024 to December 2025. Subjects who meet the eligibility criteria will be randomly assigned to either the intervention group (cesarean section + REBOA) or to traditional treatment group (control group) (cesarean section). A simple randomization method will be applied (Figure 1). Randomization will be based on a "single sequence of random assignments known as simple randomization" [27]. This technique maintains complete randomness of the assignment of a person to a particular group. The randomization procedure includes using a shuffled deck of cards (odd - intervention group, even - control group) [27]. The randomization will be performed after patients' recruitment, when the study aims and expected outcomes are disclosed to a patents and an informed consent is signed, by the responsible project team member.

Sample size calculation To estimate the required sample size, effect sizes for three major outcomes (mean intraoperative blood loss, blood transfusion volume, and hysterectomy rates) were derived from previous studies examining the impact of REBOA on reducing blood loss during cesarean sections in patients with placenta accreta spectrum disorders. Due to substantial heterogeneity in the effect sizes across studies (I2 for mean blood loss difference = 94.89% and I2 for mean transfusion volume = 86.95%), random-effects meta-analyses were conducted on individual study-level blood loss and transfusion volumes extracted from two previously published meta-analyses [28,29]. For hysterectomy outcome, a fixed-effect meta-analysis was applied due to lower heterogeneity (Q = 19.90, I2 = 29.64%).

The meta-analysis of mean blood loss differences between REBOA and control groups, based on studies [28,29], revealed a random-effects pooled estimate of -1331.46 mL (95% CI: -1643.21 to -1019.71) and a pronounced effect size (Cohen's d = - 2.06). Using the pooled mean difference (Δ_"pooled" ^ ) and variance adjusted for between-study heterogeneity (〖"SE" 〗^2+τ^2), the required sample size was calculated as follows:

n =(2×(Z_(α\/2)+Z_β )^2×(〖"SE" 〗^2+τ^2 ))/(Δ_"pooled" ^2 ) = (2×(1.96+〖0.84〗_ )^2×(〖159.06〗^2+〖292106.16〗^2 ))/(〖-1331.46〗_^2 ) = 2.81 ≈ 3 patients per group

Similarly, a meta-analysis of blood transfusion volume differences, based on 13 studies, yielded a pooled estimate of -975.10 mL (95% CI: -1224.25 to -725.95) with a similar effect size (Cohen's d = - 2.08), resulting in a required sample size estimate of 2.99 (≈ 3 patients per group). These small sample size estimates reflect the large effect sizes observed, where the pooled mean differences are substantial relative to the pooled standard deviation, yielding a high signal-to-noise ratio. This robustness persists even after accounting for high heterogeneity in continuous outcomes.

For hysterectomy rates, fixed-effect meta-analysis across 15 studies revealed a pooled odds ratio (OR) of 0.31 (95% CI: 0.21 to 0.46) with a significant overall effect (Z= -5.84, p < 0.001)

To calculate the required sample size for this outcome, the total proportions of hysterectomy events in the control groups (p_(1 ) = 0.314) and study group 〖(p〗_2 = 0.081) were incorporated as follows:

n=((Z_(α\/2)+Z_β )^2⋅[p_1 (1-p_1 )+p_2 (1-p_2 )])/(p_1-p_2 )^2 = 41.86 ≈ 42 patients per group

Using a sample size greater than the largest estimated requirement (n ≈ 42 per group) among major outcomes will ensure sufficient power for evaluating the impact of REBOA in minimizing blood loss during cesarean sections in patients with placenta accreta spectrum disorders, adequately accounting for the detection of previously reported effect sizes with a significance level (α) 0.05 and a statistical power (1-β) of 80%. Thus, the calculated number of samples size was rounded up to 50 per group (overall 100 patients).

Diagnostic methods All patients will undergo clinical and anamnestic data collection procedures (patient interview for identification of chief complaints, past medical history, and performing external obstetric examination), (Table 1).

Laboratory tests (CBC, biochemical blood test, coagulation tests, and thromboelastography) will be performed following the general examination. Thromboelastography will be carried out at the transfusiology department of UMC using the TEG5000 Haemonetics 2017 thromboelastograph. Thromboelastograph (TEG) is a device used to assess the mechanical strength of clot formation using the interaction between coagulation factors and platelets, which cannot be assessed in other laboratory tests such as coagulogram [30,31]. A blood test for thromboelastography will be taken in (1) the preoperative period (24 hours before surgery); (2) 10 minutes after fetal extraction; (3) immediately after closing the skin wound; (4) twenty-four hours after surgery. Key TEG parameters include reaction time (R), clot formation time (K), maximum amplitude (MA), alpha angle (α), percent lysis 30 minutes after MA (Ly30), and coagulation index (CI; a calculated index of total coagulation).

Assessing the risk of coagulation will help to predict thromboembolic complications in the postoperative period and recommend adequate thromboprophylaxis.

Ultrasound investigation will be used to confirm the diagnosis of PAS disorders. The main ultrasound signs of placenta accreta are: placental lacunae, disappearance of the normal hypoechoic retroplacental zone, abnormal structure of the border between the uterus and the bladder wall, pathological pattern of blood flow with color Doppler mapping. Loss of the hypoechoic retroplacental zone and subplacental hypervascularization are more common in placenta increta, while vascular lacunae and "uterine hernia" are associated with placenta percreta. Ultrasound diagnostics will be carried out using a Philips ultrasound machine, equipped with a 9.4 Hz vaginal probe and a 5.1 Hz convex probe with Doppler.

MRI of the pelvic organs without contrast to confirm the diagnosis and the degree of placenta accreta in case of placenta location on the posterior wall, suspected placenta percreta, unsatisfactory visualization with ultrasound (in obese women), also to determine angiometric parameters of the abdominal aorta before endovascular intervention. MRI diagnostics will be carried out using the PHILIPS Ingenia 3.0 T device.

Pathohistological examination of the placental and eventual resected uterine tissue will be performed after cesarean section to confirm placenta accreta spectrum syndrome [32,33]. Histological examination will confirm the diagnosis of placenta accreta by extended areas of missing decidua between the placenta and myometrium. These may include an area with placental villi affixed directly to the myometrium or abnormal implantation with a coating of fibrinoid and intervening trophoblast between the placental villi and the myometrium. In cases of placenta accreta differentiation of decidual cells and extravillous trophoblastic cells could be challenging. Myometrium is not expected to be thinned. Contrary to placenta accreta, for placenta increta cases presence of chorionic villi within the myometrium with thinned myometrium. For placenta percreta no remaining myometrium is identified, dye present on tissue attained from area suspicious for placenta percreta. Invasion through the myometrial wall is recognized in the area of percreta or villous tissue adjacent to extrauterine structures (bladder) [32,33].

Interventions REBOA. In the operating room, a patient will be positioned on the operating table (supine position). After appropriate sanitation of the surgical field, under local anesthesia S. Lidocaine 1% - 10 ml, the right common femoral artery will be punctured under ultrasound guidance. An introducer of the appropriate diameter will be installed and washed with heparinized saline. A compliant or semi-compliant balloon of the appropriate diameter will be installed into the lumen of the aorta Zone IIIa (proximal zone), (Figure 2) under ultrasound guidance. After that, a team of obstetricians will perform cesarean section and extraction of the fetus. At the time of fetus extraction, the balloon will be inflated in the lumen of the aorta with saline solution using a 20 ml injection syringe [16,34-36]. Manual control for the absence of pulsation in the common femoral arteries on both sides will be performed for control. The maximum inflation time of the balloon will be 15 minutes. If necessary, the balloon could be inflated repeatedly, with a break of 5 minutes between periods of aortic occlusion. At the end of the main stage of the operation, the balloon will be deflated step by step under the control of invasive blood pressure measurement. The introducer will be removed using suturing devices and manual compression of the puncture site for 10 minutes. An aseptic pressure bandage will be applied at the end.

Cesarean section surgery will be performed according to the standard procedure following the national guideline on "Cesarean section" used in the NRCMC [37].

Ethical considerations All potential study participants will receive an explicit explanation of the study aims, procedures, and potential outcomes. After the detailed explanations written informed consent will be obtained from patients who agree to participate in this study. The study was approved by the Institutional Ethical Board of the University Medical Center on 10.11.2023, protocol #2023/01-028.

Patient withdrawal Any study participant who no longer agrees to participate in this study can withdraw at any moment without consequences for their subsequent healthcare. Data of patients who decided to be withdrawn from the study will not be analyzed.

Statistical analysis Statistical analyses will include both descriptive and inferential methods. Baseline characteristics of the study population will be summarized as mean ± standard deviation (SD) or median with interquartile range (IQR) for continuous variables, and as frequencies and percentages for categorical variables. Inferential analyses will compare group differences in major outcomes using independent samples t-tests for normally distributed data or Mann-Whitney U tests for non-normally distributed data. Categorical outcomes will be assessed using Pearson's χ² test or Fisher's exact test, as appropriate. For multi-group comparisons, one-way analysis of variance (ANOVA) or Kruskal-Wallis tests will be employed, depending on data distribution.

To quantify the effect of REBOA while adjusting for potential confounders, regression models will be used for continuous outcomes, and classification models will analyze categorical outcomes such as hysterectomy or complications. Statistical analyses will be conducted using Python, leveraging the Statsmodels (v0.14.4), SciPy (v1.11.3), and Scikit-learn (v1.3.0) libraries for inferential statistics and modeling; NumPy (v2.1.3) and Pandas (v2.1.1) for data handling; and Matplotlib (v3.8.1) and Seaborn (v0.13.2) for data visualization.