Intraoperative Plasma Exchange for Sensitized Lung Transplantation Candidate Leads to Significant Perioperative Coagulopathy

Abstract

Sensitized lung transplantation recipients are at an increased risk for complications. Therapies such as plasma exchange (PLEX) exist to lessen the antibody burden in hopes of mitigating acute and chronic complications. Although the use of PLEX has been shown to be an effective immunomodulation for many diseases, its benefit in lung transplantation has been debated, and the fear of PLEX-induced coagulopathy has curtailed its use more broadly. Although the type of fluid used for exchange can determine the severity of coagulopathy, even the use of fresh frozen plasma (FFP) can disrupt the coagulation cascade. We present the case of severe perioperative coagulopathy following PLEX with FFP during bilateral lung transplantation.

INTRODUCTION

Transplant candidates with elevated human leukocyte antigen (HLA)–specific antibodies have been shown to have an increase in short- and long-term complications. These potential complications cause many institutions to decline donor lungs, leading to these patients being less likely to receive a lung transplant and more likely to decease while on a waiting list.[1] Accordingly, desensitization has emerged as a focus for improvement to safely include a vaster range of HLA-sensitized patients. Therapeutic options available for use in the perioperative setting include plasmapheresis or plasma exchange (PLEX), intravenous immunoglobulin (IVIG), concurrent splenectomy, and antithymocyte globulin (ATG). Although there are variations among institutional protocols, perioperative PLEX is a foundational component of many strategies.[2]

Plasma exchange involves removing a patient’s plasma and exchanging it with a replacement fluid. The choice of replacement fluid will vary but often consists of albumin, fresh frozen plasma (FFP), or a combination of the two.[3] Although the use of PLEX can be an effective option for desensitization, it is an invasive procedure with its own risks of complications. PLEX can increase perioperative blood loss through a reduction in fibrinogen and platelet levels and—depending on the type of fluid used for exchange—a significant removal of coagulation factors.[4,5] This increased risk of bleeding must be weighed against additional aspects of lung transplantation that are associated with perioperative blood loss and can present a clinically challenging circumstance for transplant teams to navigate. We present the case of a highly sensitized adult woman who underwent bilateral orthotopic lung transplantation (BOLT) with intraoperative PLEX complicated by significant perioperative bleeding.

CASE HISTORY

This case entails a 31-year-old woman with a history of end-stage interstitial lung disease who underwent BOLT. Pertinent medical history included pulmonary hypertension, chronic steroid use, and normal biventricular function on transthoracic echocardiogram. Antibody screening revealed an elevated panel reactive antibody (PRA), and Hematology was consulted for potential PLEX. Following evaluation and coordinated discussion with the transplantation team, PLEX therapy was elected based on the current apheresis guidelines for HLA desensitization and lung transplant as a category III grade 2C recommendation.[6] The first session was to be completed in the operating room followed by 2 sessions postoperatively for a goal of 1.5 L volume exchange for each session.

The patient arrived at the operating room in a critical but stable condition. General anesthesia was induced uneventfully. An arterial blood gas (ABG) and point-of-care (POC) viscoelastic test with Quantra (Quantra Hemostasis Analyzer, HemoSonics, Durham, NC) were obtained prior to incision with no significant abnormalities [Table 1]. Intraoperative PLEX was organized and managed by our transplant team. A total of 16 units of FFP were administered to achieve an exchange of 1.5 times the patient’s plasma volume. This was coordinated with the anesthesia team, and the patient’s hemodynamics were continuously assessed. Hemoglobin and calcium concentrations were monitored with ABGs and corrected as needed. Roughly 6 grams of calcium chloride were administered throughout the process. PLEX was initiated upon incision and completed prior to central cannulation for venoarterial extracorporeal membrane oxygenation (VA ECMO).

Table 1.

Intraoperative point-of-care viscoelastic testing using quantra

	Clotting time (seconds)	Clot strength (hPa)	Fibrinogen contribution to stiffness (hPa)	Platelet contribution to stiffness (hPa)
Normal parameters	104-166	13-33.2	1-3.7	11.9-29.8
Using Quantra
Prior to incision	148	32.8	4.5	28.3
Following plasma exchange therapy		8.2	0.8	7.4
Protamine administration and VV ECMO initiation	225	4.7	0.5	4.2
Following PCC administration	142	17.5	2.7	14.5

hPa=hectopascals, PCC=prothrombin complex concentrate, VV ECMO=venovenous extracorporeal membrane oxygenation

After the initiation of VA ECMO the recipient’s right lung was prepared for pneumonectomy. It was during this point of the operation that the patient required extensive resuscitation with intravenous fluids and packed red blood cells (pRBCs). At this time, a repeat Quantra was performed and showed a significant change from baseline [Table 1]. One unit of platelet apheresis and one unit of cryoprecipitate was administered as the right lung was reperfused uneventfully. The recipient’s left lung was dissected and transplanted in a similar fashion and reperfused without any significant issue.

After bilateral transplantation, we proceeded to wean from VA ECMO while oxygenation was assessed with the use of pulse oximetry and biventricular function assessed using TEE. Hemodynamically, the patient tolerated weaning with no change in pressor requirement, but a significant drop in the patient’s oxygenation was noted on the pulse oximeter and confirmed with PaO2 on ABG. After adjusting our ventilator settings, we were unable to wean flows beyond 3 L/min and the patient was transitioned uneventfully to venovenous extracorporeal membrane oxygenation (VV ECMO).

During chest closure, the patient remained significantly coagulopathic and required continual resuscitation despite a return of activated clotting time to baseline following protamine administration. A third Quantra, obtained after conversion to VV ECMO and protamine administration, demonstrated a continual coagulopathy [Table 1]. In addition, our clotting time ratio was within normal range (1.3) indicating appropriate reversal of heparin. Additional resuscitation with platelets, cryoprecipitate, and FFP ensued. In total, our patient received 3 units of platelets apheresis, 3 units of cryoprecipitate, 3 units of FFP, 9 units of pRBCs, and 3.2 L of cell saver yet remained coagulopathic as noted by our surgical colleagues and upon inspection of bilateral chest tube output. After an intraoperative discussion between our anesthesia team, surgical colleagues, and perfusionist, it was decided to administer 25 units/kg of prothrombin complex concentrate (PCC) and trend coagulation with repeat Quantra. Flows for VV ECMO were kept at a minimum of 4 L/min to help mitigate any potential risk of machine thrombosis. In addition, a second ECMO machine was readily available in the unit in the event thrombosis were to become an issue.

After the administration of PCC, a decrease in bilateral chest output was noted and repeat Quantra showed an improvement in our coagulation profile [Table 1]. The chest was closed, and the patient was transported to the ICU. After arrival, the patient was continually resuscitated, and she was ultimately brought back to the operating room for chest exploration. The patient’s postoperative course was complicated by mild to moderate antibody-mediated rejection (AMR) (A2B0C0) and VV ECMO requirement. Postoperative PLEX was deferred due to patient instability. Her hospital course spanned 114 days, but she was ultimately discharged home on room air.

DISCUSSION

Patients with anti-HLA antibodies have been shown to be at an increased risk for acute cellular rejection, CLAD and AMR. These feared complications have ultimately led to many institutions declining donor organs for sensitized patients.[1] Historically, PLEX has been the standard immunomodulation for many diseases, but its effectiveness for sensitized patients undergoing lung transplantation has been debated. The Apheresis Applications Committee of the American Society for Apheresis groups plasmapheresis for sensitized lung transplantation recipients into category 3 of 4, which concludes there is minimal evidence for the benefit of PLEX and therapy should therefore be individualized.[7]

The severity of PLEX-induced coagulopathy will vary depending on the type of fluid used for exchange. The use of a colloid solution, such as albumin, is associated with lower rates of viral transmission, allergic reactions, and more readily available but can increase the risk of bleeding owing to an absence of clotting factors. The use of FFP for exchange therapy is believed to mitigate the risk of dilutional coagulopathy but increases risk of both transfusion related reactions and viral transmissions.[8] However, even the use of FFP for exchange therapy can increase the risk of bleeding from an overall reduction in platelets and to a lesser extent fibrinogen. FFP contains coagulation factors in therapeutic amounts, but it lacks platelets and in large amounts can have a dilution effect on a patient’s platelet concentration.[9,10] Similarly, although FFP does contain fibrinogen, if the fibrinogen concentration replaced does not match the fibrinogen concentration removed, the patient can develop a reduction in fibrinogen levels.[10] The coagulation cascade is a delicate process, and PLEX therapy, even with FFP, can disrupt that process. While there is evidence that the use of intraoperative PLEX may increase perioperative bleeding, it is difficult to quantify the extent, and this has ultimately curtailed the use of PLEX more broadly during lung transplantation.[4]

The use of ECMO for lung transplantation will vary by institution and patient, but its use can further add to the complexities and the risks of intraoperative PLEX. The use of ECMO significantly increases the odds of bleeding, as shown by Adelmann et al.[11] who found that preoperative ECMO was associated with increased odds of severe postoperative hemorrhage (OR = 14.41%, P < 0.001). Although the transfusion of blood products for patients on ECMO has been safely described in the literature, the use of PCC is scarcely cited. Presently, there are no studies on the effectiveness and safety of PCC administration in adult patients on ECMO.[12]

While our institution remains a high-volume lung transplantation center, the use of perioperative PLEX for desensitization is infrequently used. However, given the routine use of intraoperative ECMO during transplantation at our institution, the decision to use FFP for exchange therapy was sought in the hope of mitigating any increased risk of perioperative bleeding. The degree that PLEX contributed to our patient’s coagulopathy is unclear but given our patient’s normal coagulation profile prior to the procedure, the significant change in our POC viscoelastic testing following exchange therapy, and the absence of any significant surgical bleed, we felt that intraoperative PLEX had some role in our significant blood loss. While transfusion thresholds differ amongst various centers, the use of PCC during lung transplantation at our institution is rarely considered and requires a real-time discussion among team members of the risks and benefits of its use.

This case highlights the challenges of lung transplantation for patients with anti-HLA antibodies and showcases the potential coagulopathy that intraoperative PLEX can present, especially when combined with additional factors which can increase bleeding during lung transplantation. Given the lack of a definitive benefit of PLEX for lung transplantation, as stated by the American Society for Apheresis, the risk of increasing bleeding during a surgical procedure that is already high risk for blood loss must be individualized. In our instance, excessive blood loss led to the administration of PCC for a patient dependent on ECMO, and the patient developed AMR despite intraoperative exchange therapy.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient (s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initial s will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.