Prophylactic endovascular balloon occlusion of the aorta in cases of placenta accreta spectrum during caesarean section: points from the anaesthesiologist’s perspective

In this study, we found that balloon occlusion could reduce EBL and the ratio of EBL ≥ 1000 ml. Here, in patients with PAS, PAABO was confirmed to have more advantages in reducing blood loss during caesarean delivery than traditional interventional therapies. These results were consistent with previous research [11].

In response to this novel technology, it is imperative for anaesthesiologists to be familiar with the physiologic changes induced by balloon inflation and deflation to conduct appropriate anaesthetic methods and to address massive haemorrhage. Until now, few reports have addressed the points about anaesthetic management. Anaesthesiologists’ goal is to develop an appropriate anaesthesia regimen with no adverse effects on women and babies, to minimize fluctuations during operative haemodynamics, to address haemorrhagic events, and to prevent perioperative complications [11].

Anaesthesia for CS with PAS remains controversial. Regarding the potential impact of anaesthesia on babies, the preferred anaesthetic method for CS is regional anaesthesia [12]. However, GA is superior to RA in patients with a risk of severe haemorrhage or prolonged surgical time. According to the previous literature, there is no significant difference in neonatal outcomes between regional and general anaesthesia [12]. In the early cases, we attempted to operate under RA but failed due to the severe discomfort of the parturients with abdominal distension at balloon inflation. Therefore, we selected GA for the remaining 24 parturients in Group A. For the parturients in Group B, we preferred RA, with 20 parturients in Group B receiving RA and 3 parturients receiving GA due to an emergency or haemorrhagic shock. In GA, the amount of anaesthetic crossing the placenta into the foetal circulation is determined by the physicochemical and structural properties of the drug as well as the dose and duration of transplacental transfer. Therefore, the duration from infusion of propofol and remifentanil to delivery of babies should be as short as possible. However, abdominal adhesion and prolonged induction-to-delivery interval might commonly occur in parturients with PAS due to a history of CS. Thus, to minimize adverse effects on babies, experienced anaesthesiologists and obstetricians as well as intimate teamwork are needed. In this study, the average duration of induction-to-delivery was 6.6 ± 2.1 min, which was longer than common CS. However, the skin incision-to-delivery interval was 5.4 ± 1.8 min in the GA group, which was significantly shorter than that in the RA group (8.0 ± 2.7 min). The difference might be attributed to the more experienced anaesthesiologists and obstetricians with intimate cooperation in the GA group. In this study, there were no differences in the Apgar scores when comparing Group A with Group B or GA with RA.

The results showed that arterial blood pressure was significantly elevated after balloon inflation, which was consistent with the report by Xin Wei [13]. A modest increase in arterial blood pressure was helpful for maintaining perfusion of the heart and brain, but excessively high pressure might be hazardous to patients with potential cardiac, pulmonary, or cerebral disease. A sudden elevation in arterial pressure would increase afterload and left ventricular work, which could lead to left ventricular dilation, myocardial ischaemia, and cerebral hyperperfusion. This might induce intracranial trauma or bleeding [14]. A recent report showed that increased blood pressure induced by balloon inflation can increase cerebral blood flow risk and exacerbate a pre-existing cerebral injury such as a contusion or intracranial haemorrhage [15]. Current guidelines recommend an optimal cerebral perfusion pressure (CPP) of 60 mmHg for patients during balloon inflation [14]. In addition, there would be a drastic reduction in blood pressure when the balloon was deflated, which might lead to ischaemia of the myocardium, brain, or intestine. Therefore, it is important for anaesthesiologists to take measures to reduce fluctuations in haemodynamics during balloon inflation and deflation. Relevant measures include increasing the depth of anaesthesia and the use of opioid analgesics and vasoactive drugs. IBP monitoring is imperative for real-time observation of vital signs and monitoring of blood gas indicators.

Complications are the most important features that anaesthesiologists need to focus on with regard to PAABO. The relevant complications include embolization of the lower extremities [7], ischaemic injury of the myocardium, kidneys, lungs, intestines and spinal cord [16] and aorta rupture or dissection, which can lead to maternal death [17]. To prevent or reduce the potential risks of PAABO, the following should be performed: (1) minimization of the occlusion time,(2) maintenance of the balloon diameter at the smallest size that blocks blood flow and (3) localization of the balloon below the renal artery origin to reduce the risk of renal injury. As currently there is no uniform standard for the duration of balloon occlusion [11]. Yeung and Wahlberg [16, 18] showed that the morbidity of postoperative renal failure decreased at < 20 min of ischaemia and increased 10-fold at > 50 min of the ischaemia. However, some studies have shown that the spinal cord might tolerate the occlusion up to 30 min [19, 20]. Andoh et al. [21] suggested a 25-min occlusion time in caesarean hysterectomy for placenta accreta. Masamoto et al. [22] suggested that the longest occlusion time be 80 min without intra- or postoperative complications. The maximum occlusion time for parturients needs to be explored. To prevent or reduce the risk of aortic rupture, the occlusion time and diameter of the balloon should both be restricted [23]. Regarding the diameter of the balloon, Liu et al.. confirmed that the diameter of the aorta could be estimated by pelvic MRI prior to surgery and that the commonly used size was 16 ~ 20 mm [1]. In this study, the longest continuous duration of balloon inflation was restricted to less than 15 min. If a continuous long-term block is needed for surgery, the balloon inflation should be alternated with 1-min deflations to restore the blood supply [17].The principle of balloon expansion in our operation was to use the minimum volume to block blood flow and limit the maximum volume to 15 ml. The mean volume for inflation of the balloon was 9.9 ± 3.4 ml. The results showed that the rate of neurologic complications of the lower limbs in Group A was 8%. Gastrointestinal complications were found in 14 cases (10 in Group A and 4 in Group B). The most common gastrointestinal dysfunction in Group B was abdominal distension, which might be a result of low intestinal movement after GA. Meanwhile, two more severe cases with paralytic intestinal obstruction (1 case in Group B) as well as upper gastrointestinal stress ulcer (1 case in Group A) might due to gastrointestinal stress response after severe bleeding. Hence, for the patients with massive bleeding, the proton pump inhibitors should be administered to reduce the stress injury to the gastrointestinal mucosa. The index values of GPT and GOT significantly increased in both groups when comparing the preoperative and postoperative indicators. The indexes were still in the normal range, suggesting that the PAABO operation did not lead to severe gastrointestinal complications. The index of postoperative LDH activity was significantly elevated in both groups, and no significant differences were found between the groups. LDH is one of most important glycolytic enzymes in cells, and it functions as a marker of ischaemia/reperfusion injury. LDH is released into the blood when cells are damaged. Haemorrhage and ischaemia/reperfusion injury can elevate LDH activity and lead to gastrointestinal/intestinal complications [24]. All of the complications and elevated biochemical indexes were resolved and returned to normal, respectively, and there was no permanent harm.

The present study had several limitations. First, it was a retrospective and single-centre study, in which there was a lack of randomness in grouping, leading to selection bias. Second, the sample size was relatively small, which might lead to a less statistically powerful sample and unstable and inaccurate results. Third, we compared the effect of GA with RA in all patients with PAS rather than those with PAABO, resulting in some defects of imbalance and incompletion. Finally, the depth of the study is insufficient, and some indexes such as lactic acid could have been affected by inflation balloon. Therefore, a prospective and multicentre study with more patients is needed.