Intraoperative extracorporeal membrane oxygenation (ECMO) is routinely used during lung transplant surgery to provide the patient with temporary respiratory and circulatory support [1–3]. Over the past two decades, many centres have switched from intraoperative support with cardiopulmonary bypass (CPB) to ECMO [1, 2, 4, 5]. The intraoperative ECMO approach allows surgeons to perform the procedure with greater precision and efficiency while minimizing the risk of lung injury [2, 4]. Most centres that use intraoperative ECMO, to a greater extent, support the idea that ECMO helps to control lung graft reperfusion [4, 6, 7]. Another advantage of intraoperative ECMO support is the possibility of transplanting more patients with comorbidities who would not be able to undergo surgery on selective unilateral ventilation without extracorporeal support. For example, patients with severe pulmonary hypertension have a high risk of left-sided heart failure, pulmonary fibrosis, and very low pulmonary compliance [8, 9]. ECMO has a much lower pro-inflammatory potential than CPB [1]. However, ECMO also carries risks of bleeding, infection, and thrombotic complications [10–13]. The entire team should decide to use intraoperative ECMO on a case-by-case basis, considering the patient’s medical history and condition [14, 15]. Our study aims to highlight that lower doses of anticoagulants administered during intraoperative ECMO support do not lead to increased risks and may be beneficial for the patient.

This study was approved by the Local Ethics Committee (reference number EK-786/23) and was registered in the clinical trial database at ClinicalTrials.gov (identifier number NCT06054997). This study was designed as a single-centre, retrospective, observational study that included all lung transplants performed between January 2020 and December 2022 within the Prague Lung Transplant Program Motol University Hospital. A total of 141 patients underwent transplantation during the study period. The exclusion criteria were lung transplantation without intraoperative ECMO support, block heart-lung transplantation, ECMO bridge-to-lung transplant, and planned postoperative ECMO. The inclusion criteria were lung transplantation performed with intraoperative ECMO support, central ECMO cannulation, and successful ECMO support termination at the end of surgery. According to the study inclusion criteria, only lung transplants performed under intraoperative central ECMO cannulation with successful ECMO weaning were included. In total, 109 patients fulfilled the inclusion criteria. Thirty-two patients were excluded based on the following criteria: heart-lung transplant (HLTx), n = 4; ECMO bridge, n = 4; transplantation (Tx) without ECMO, n = 8; and planned prolonged ECMO, n = 16. UFH was used for ECMO anticoagulation in all the patients. The subjects were divided into two groups for further analysis according to the UFH dose administered during the entire surgical procedure. In the first group, the UFH dose was ≤60 IU/kg/surgery. In the second group, the UFH dose was greater than 61 IU/kg/surgery. A cutoff value of 60 IU/kg was determined based on the available literature review. Values ≤60 IU/kg/surgery were considered relatively lower doses, and values >61 IU/kg/surgery were considered higher doses of UFH [2, 4, 10–13, 16–19]. The UFH effect was monitored using activated clotting time (ACT) values. The intraoperative haemoglobin level target for red blood cell (RBC) substitution was 100 g/L. Coagulopathy was managed according to the clinical experience of the anesthesiologist and viscoelastic Point of care methods (ROTEM, PFA). The parameters followed up intraoperatively in both groups were total blood loss in milliliters and related to the patient’s weight during the surgery (assessed by the amount of blood in a calibrated suction device); the total amount of UFH administered to the patient during surgery in the international unit (IU) and related to patient weight; and the consumption of blood derivatives during the surgical procedure, such as RBC, fresh frozen plasma (FFP), and platelets (PLT). In both groups, ACT values were monitored after the administration of UFH before ECMO cannulation and then every 60 min. In both groups, protamine was administered at the end of the surgical procedure until physiological ACT values below 120 s were achieved (if needed). No type of biological glue to seal the anastomosis was used. In both groups, intraoperative VA ECMO was implanted electively, and no patient underwent urgent ECMO cannulation due to cardiac or pulmonary reasons. The Maquet Rotaflow RF-32 centrifugal pump provided Intraoperative VA ECMO support. Heparin-coated cannulas and a heparin-coated tubing system were used for cannulation. According to internal guidelines, the ECMO flow was maintained at 1/2 to 2/3 of the calculated cardiac output. In the postoperative period, we followed up on the development of hemothorax requiring surgical revision. We considered surgical revision for hemothorax to be a significant bleeding complication. We also evaluated blood product consumption in the period 48 h after ICU admission, PGD grade three in 72 h after LUTx, 30-day and 90-day mortality, long-term survival and ECMO circuit-related and patient-related thrombotic complications. The basic, standard points of care in the Intensive care unit (ICU) are displayed in Table 1.

Patients were recruited between January 2020 and December 2022, and based on the exclusion criteria, a total of 32/141 patients were excluded from the study. A flow diagram based on the Consolidated Standards of Reporting Trials (CONSORT) is shown in Figure 1. The lower dose of UFH (≤60 IU/kg/surgical procedure) and higher dose of UFH (>61 IU/kg/surgical procedure) groups ultimately consisted of 44 and 65 patients, respectively. For most parameters, we obtained surprising results, clearly in favour of the administration of lower doses of UFH (≤60 IU/kg). Total blood loss during surgery was significantly lower in the group treated with lower doses of UFH (≤60 IU/kg). The mean total intraoperative blood loss was 753 and 1,470 mL, respectively (p < 0.0001) (Table 3). Blood loss related to body weight was also significantly lower in the UFH group ≤60 IU/kg). The mean intraoperative blood loss/patient body weight was 9.628 mL/kg and 20.97 mL/kg, respectively (p < 0.0001) (Table 3). The total UFH dose was significantly lower in the UFH group (≤60 IU/kg). The mean total intraoperative UFH doses were 3491 IU and 8694 IU, respectively (p < 0.0001) (Table 3). The total UFH dose, based on body weight, was significantly lower in the UFH group (≤60 IU/kg). The mean dose of UFH/patient bodyweight intraoperatively was 43.81 IU/kg and 116.6 IU/kg, respectively (p < 0.0001) (Table 3). We also noticed a significant difference in favour of reducing the consumption of blood derivatives in the group with lower doses of UFH ≤60 IU/kg. The consumption of RBCs during the surgical procedure was 0.5581 and 1.908 units, respectively (p = 0.0009) (Table 3). The FFP consumption during surgery was 0.4186 and 1.862 units, respectively (p = 0.0009) (Table 3). The platelet consumption during surgery was 0.1628 and 0.4154 units, respectively (p = 0.1461) (Table 3). The mean ACT values before ECMO cannulation and 3 min after the administration of UFH were lower in the UFH group ≤60 IU/kg (156.3 and 209.1) (p < 0.0001) (Table 3). There was a significant reduction in bleeding complications in terms of surgical revision for hemothorax in the lower UFH dose group ≤60 IU/kg, with only one revision (2.27%) and nine revisions (13.85%), respectively (p = 0.040) (Table 3). In the postoperative period 48 h after admission to the ICU, we did not observe a significant decrease in the consumption of blood derivatives (Table 4). However, there was a significantly lower incidence of third-degree PGD 72 h after LUTx in the group where a lower dose of UFH was administered (p = 0.038) (Table 4; Figure 2). The 30-day, 90-day, and long-term survivals were not different (Figure 3). The log-rank test was p = 0,6879 (Figure 3). Mortality rates were not different in either group. We did not record any thrombotic complications arising from the ECMO circuit or patient-related complications in any group.

Intraoperative ECMO support during lung transplantation is a routine method frequently used to facilitate surgical procedures. This enables the procedure to be performed in significantly more polymorbid patients who cannot handle selective ventilation. Another indisputable advantage of ECMO is the possibility of controlled reperfusion of lung grafts. During ECMO support, the blood is in contact with allogeneic materials such as cannulas, circuits, and oxygenators. All of these factors can cause potential complications, including bleeding and thrombosis. Therefore, balancing the edge between anticoagulation and procoagulation is very important to minimize complications during intraoperative ECMO support. UFH remains the most widely used drug for ECMO anticoagulation. From the available recommendations of thoracosurgery societies, we know the recommendation for UFH dosing, most often between 25–100 IU/kg and effect control with ACT in the range of 180–220 s. However, the literature and the guidelines more frequently report the possibility of reducing the dose of UFH and, very importantly, without an increase in thrombotic complications. According to the recommendations of the thoracosurgery societies, in the case of bleeding complications or anticipated intraoperative bleeding (for example, significant intrapleural adhesions), it is recommended to minimize the dose of UFH or completely eliminate UFH and perform heparin-free ECMO. For example, Bernhardt et al. mentioned in The International Society for Heart and Lung Transplantation/Heart Failure Society of America Guidelines on Acute Mechanical Circulatory Support that “bleeding complications in acute mechanical circulatory support (MCS) are common and frequently necessitate withdrawal of anticoagulation” and stated that “in the settings of life-threatening bleeding, full discontinuation of all anticoagulation may be necessary” [20]. Hartwig et al. stated in The American Association for Thoracic Surgery guidelines that “low or no heparin regimes are recommended for patients with significant adhesions and impaired coagulation status” [1]. Additionally, Lorusso et al., in the EACTS/ELSO/STS/AATS expert consensus, mentioned that “anticoagulation is required during prolonged ECLS to prevent circuit thrombus formation with embolization and/or circuit failure. However, bleeding remains the most frequent complication associated with ECLS. UFH infusion is typically delayed until haemostasis is achieved, often within 24–48 h. Reports that suggest the safety of prolonged withdrawal of anticoagulation for as long as 3 days when faced with bleeding are important” [21]. This trend was also confirmed by our study, which suggests that a lower dose of UFH as part of intraoperative ECMO support may not pose a risk but, on the contrary, can have substantial benefits for the patient. Using lower doses of UFH below 60 IU/kg (43.81 IU/kg based on our study) can benefit the patient and the entire perioperative period. Nonetheless, these results must be interpreted with caution, and numerous limitations should be considered. The major limitation is that the dosage of the UFH used in the study (high or low dose) has been based on subjective criteria, i.e., the clinical experience of the anesthesiologist guides its dosage. Another possible limitation of this study may be the division of subjects according to UFH into groups below and above 60 IU/kg since the available literature does not define the exact dose of UFH but only the range. Therefore, it was necessary to set a limit for dividing the patients into groups. However, the medians of the two groups were very far apart 43.81 IU/kg vs. 116.6 IU/kg; we presume this is more of a minor bias. Furthermore, having a more significant number of investigated subjects would be advisable, which would add even more weight to the entire study. Similarly, a particular bias may have been introduced into this study because the primary hemostasis disorder was not investigated; primary hemostasis can be disturbed as part of ECMO support during lung transplantation, thereby potentiating intraoperative bleeding [22]. The authors hypothesize that the significantly higher incidence of PGD grade 3 in 72 h postop in the group with higher doses of UFH may be caused by the need to administer a larger number of blood derivatives, which is one of the possible reason for the development of PGD. The anesthesiologist plays a crucial role in the decision-making process and is mostly responsible for anticoagulation management of intraoperative ECMO support [15, 23, 24]. Such management depends not only on their experience but also on their knowledge of the latest findings and recommended practices. The decision to withhold anticoagulation therapy or decrease its dosage involves balancing the competing risks between bleeding and clotting.