The Year 2020 in Review: Noteworthy Literature for Thoracic Transplant Anesthesiologists

Preoperative Evaluation and Risk Factors

Patient selection and preoperative optimization are essential components for promoting successful outcomes after LTx. Hernandez-Morgan et al ¹ performed a single-center, retrospective analysis to assess the impact of preoperative anemia on outcomes for LTx. Out of 342 patients included in the cohort, 54% (n = 183) were anemic (hemoglobin <13 g/dL for males and <12 g/dL for females) prior to surgery. The primary outcome was hospital length of stay, and lower preoperative hemoglobin levels predicted a statistically significant longer duration of hospitalization (Spearman coefficient = −0.117; P = .032). The anemic group also demonstrated a higher likelihood for requiring vasopressin infusions (66% vs 55%; P = .031), a greater likelihood of requiring intraoperative transfusions (35.5% vs 18.9%; P = .001), and a 4-fold greater risk of requiring reoperation for bleeding (odds ratio [OR] = 4.89 [95% confidence interval (CI) = 1.53-15.60]; P = .007). Notably, only 11% of anemic patients received preoperative treatment for anemia, such as erythropoietin or iron supplementation, highlighting a potential area for future study in this field.

Another important preoperative consideration prior to LTx is the presence of coronary artery disease (CAD). This comorbidity can pose major challenges for LTx management, not only due to the likelihood of perioperative myocardial stress and hemodynamic instability but also because postoperative immunosuppression may accelerate the atherosclerotic process. Kanaparthi et al ² performed a single-center, retrospective analysis to evaluate the impact of prior or perioperative coronary revascularization on survival in LTx recipients. The cohort (n = 468) was divided into 4 groups: preoperative percutaneous intervention (n = 34), preoperative coronary artery bypass grafting (CABG; n = 25), concomitant CABG during LTx (n = 29), and patients undergoing LTx with no indication for coronary revascularization (n = 380). There was no difference in short- or long-term survival out to 5 years between the groups, despite the fact that the group not requiring revascularization was younger than the rest (P = .001). Of note, patients presenting with prior CABG were more likely to receive a single lung transplant on the contralateral side to the mammary graft (21 vs 4; P = .054), largely due to concerns of injuring the graft during dissection. And finally, all concomitant CABGs were performed on beating hearts without the administration of cardioplegia. Based on this work, it appears that preoperative CAD requiring revascularization does not increase mortality after LTx and, therefore, should not preclude eligibility for listing.

Similar to CAD, renal dysfunction (RD) is a common comorbidity to consider during preoperative evaluation for LTx. Woll et al ³ analyzed the UNOS (United Network for Organ Sharing) database for adult patients undergoing LTx between 1995 and 2014 to evaluate the potential impact of preoperative RD on mortality. They defined preoperative RD as an estimated glomerular filtration rate less than 60 mL/min/1.73 m² at the time of surgery. The patients with preoperative RD (n = 1337, 5.42% of the cohort) demonstrated significantly higher mortality rates at 1 year (23.2% vs 15%; P < .001) and at 3 years (38.3% vs 28%; P < .001) than patients with normal preoperative renal function. A subgroup analysis was also performed for patients who underwent simultaneous lung-kidney transplantation (SLuKi; n = 38). Despite the fact that the Lung Allocation Score (LAS) for the SLuKi group was significantly higher—suggestive of an overall higher-risk group of patients—there was no difference in 30-day, 1-year, or 3-year mortality compared with those with RD who underwent LTx alone. While this may support the utility of SLuKi in “sicker patients” with preexisting RD undergoing LTx, it is important to note that survival in the SLuKi group still remained lower than for those without preoperative RD. Preoperative RD appears to be a significant predictor of mortality after LTx, though concurrent renal transplantation may partially ameliorate the increased risk that patients with preoperative RD and a high LAS face.

Vahidy et al ⁴ sought to evaluate the impact of pretransplant opioid use on survival after LTx, as similar analysis in other transplant populations has demonstrated higher rates of graft failure and death, including kidney and liver recipients. Through retrospective analysis of adult patients undergoing LTx at their institution between 2004 and 2015 (n = 425), the group identified a preoperative prevalence of opioid use of 14% (n = 61) at the time of listing, with a median daily oral morphine equivalent dose of 31 mg (range = 18-54 mg). Overall, there was no difference in the risk of death or retransplantation between the groups (hazard ratio [HR] = 1.12 [95% CI = 0.65-1.83]; P = .657). The opioid group did require a longer hospitalization (35 days vs 27 days; P = .014), which the authors speculated to have been the result of challenges in postoperative pain management. There were no differences in any of the other secondary outcomes assessed, including duration of mechanical ventilation, intensive care unit length of stay, continued postoperative opioid use at 1 year, and time to onset of chronic lung allograft dysfunction, to name a few. This data do not support preoperative opioid use as an exclusion criteria for LTx, although it is important to emphasize the relatively low oral morphine equivalent dose range of the cohort.

Overall, evidence from this past year suggests that neither preoperative CAD nor opioid usage negatively affects survival outcomes after LTx. Meanwhile, preoperative anemia predicts greater intraoperative transfusion requirements and the risk of reoperation for bleeding, and preoperative RD is associated with increased mortality rates. Meticulous preoperative assessment and recipient selection are key elements for promoting successful LTx outcomes.

Transfusion Medicine

The perioperative management of bleeding and coagulopathy during LTx remains particularly challenging. Hemorrhage and coagulopathy are unfortunately not uncommon during LTx, but at the same time, the administration of blood products has been associated with worse outcomes, including an increased risk of primary graft dysfunction (PGD) and mortality. This year, Pena et al ⁵ published “Perioperative Management of Bleeding and Transfusion for Lung Transplantation,” which provides an excellent review of the perioperative utilization of blood products and factor concentrates during LTx. It includes the risks and benefits of transfusions, and also highlights the tremendous need for further investigation into these areas. Additionally, multiple groups published primary literature on the topic.

For example, Seay et al ⁶ contributed a retrospective, observational study of patients undergoing LTx at a single center to assess the impact of higher fresh frozen plasma (FFP) to red blood cell (RBC) transfusion ratios, which has been shown to reduce overall transfusion requirements in other patient populations. Here, only patients receiving greater than 4 units of RBCs within the first 72 hours of surgery were included (n = 89). These patients were divided into 2 groups by transfusion ratios: high-ratio group (>1:2 units of FFP:RBC; n = 38) and low-ratio group (<1:2 units FFP:RBC; n = 51). The high-ratio group received overall more transfusions (including total units, FFP, cryoprecipitate, and platelets), as well as factor concentrates. There was a much higher incidence of severe PGD at 72 hours in the high-ratio group (n = 23, 60.5%, vs n = 12, 23.5%; P = .0013), which translated to more than 3 times greater risk overall (OR = 3.33 [95% CI = 1.47-7.52]; P = .0039). On review of secondary outcomes, the high-ratio group also demonstrated significantly greater requirements for postoperative extracorporeal membrane oxygenation support and mechanical ventilation at 72 hours, as well as a longer duration of hospitalization. Despite benefit in other, non-LTx patient populations, high-ratio FFP:RBC transfusion approaches may be harmful in patients undergoing LTx.

Similarly, Huddleston et al ⁷ investigated the impact of RBC transfusion requirements, with or without recombinant human activated factor VII (rhFVIIa) on midterm survival rates after LTx (out to 6 years posttransplantation) through a retrospective, single-center analysis (n = 265). The cohort was divided into 3 groups based on specific transfusions within the first 24 hours postoperatively: no RBC transfusion (n = 78), RBC transfusion but no rhFVIIa administration (n = 149), and RBC transfusion along with rhFVIIa (rhFVIIa dose 90 µg/kg; n = 38). Overall, there was a trend toward increased mortality for patients receiving any RBC transfusion, but this did not reach statistical significance (HR = 2.168 [95% CI = 0.978-4.805]; P = .057). The subgroup that received large volumes of transfusion (>15 units RBCs) demonstrated higher mortality rates (HR = 1.363 [95% CI = 1.137-1.633]; P = .001). The cohort that was administered rhFVIIa received more than twice the volume of RBC transfusions, but after correcting for effect of these transfusions, rhFVIIa did not add any additional mortality risk for this group (HR = 1.104 [95% CI = 0.459-2.651; P = .825). Thromboembolic complications were not assessed, but overall, these data suggest that the use of rhFVIIa to correct coagulopathy during LTx may be a safe intervention that deserves further investigation.

Institutional Variations: Case Volume and Anesthetic Practices

Differences in perioperative management and outcomes after LTx between institutions were studied across a variety of metrics. Notably, the impact of case volume at LTx centers was consistently demonstrated. A retrospective cohort analysis by Jawitz et al ⁸ analyzed the association between annualized LTx volume by center with the rates of grade 3 PGD at 72 hours posttransplant, as well as survival after the development of PGD. The cohort included 7322 LTx recipients from multiple centers across the United States, 21% (n = 1525) of whom met the specific PGD criteria. After adjusting for other risk factors for PGD, the group demonstrated that increasing annualized LTx volume predicted lower rates of PGD (OR = 0.94 per 10 transplants [95% CI = 0.89-0.99]; P = .014); this association demonstrated an essentially linear relationship as LTx volume increased. Further analysis showed that the patients who did develop severe PGD had lower mortality rates as transplant volumes increased across low- and medium-volume centers (up to 55 transplants per year; HR = 0.87 per 10 transplants [95% CI = 0.79-0.94]; P < .001), although the authors note that their survival analysis was limited by a “relatively short” follow-up period.

Outcomes favoring high-volume centers were also demonstrated through other metrics across the perioperative period, including both the preoperative and postoperative settings. For example, Ranganath et al ⁹ found a long-term survival benefit in patients who required pretransplant hospitalization—an indicator of acute decompensation—if they were admitted to high-volume institutions (>25 cases per year). Postoperatively, failure to rescue (FTR) outcomes were better at high-volume institutions (>25 cases per year) according to work by Osho et al, ¹⁰ who demonstrated an independent relationship between case volume and lower rates of FTR (P < .001). Overall, these studies suggest that minimum thresholds for case volumes might be considered for LTx centers to maintain safe, high-quality care.

Beyond case volumes, differences between LTx centers were also assessed through the lens of anesthetic management patterns. Subramaniam et al, ¹¹ with support from the Society for the Advancement of Transplant Anesthesia, administered a survey of perioperative management questions to investigate differences in institutional practice patterns during LTx. Survey participants represented an international, geographically diverse cohort of members of the Society of Cardiovascular Anesthesiologists (n = 127 institutions). Topics assessed included fluid resuscitation and transfusion practices, hemodynamic monitoring and support, transesophageal echocardiography, and mechanical circulatory support, to name a few. The authors identified significant heterogeneity across a range of metrics, including restrictive practices for fluid and blood product administration, as well as inhaled oxygen concentrations at the time of reperfusion. Overall, there was limited consistency between institutions across most topics assessed, highlighting a significant need for further research to identify best practices in the perioperative management of LTx. On a similar note, work published this year by Martin et al, ¹² “The Impact of Anesthetic Management on Perioperative Outcomes in Lung Transplantation,” provides a thorough review of the available data for which anesthetic management—including preoperative, intraoperative, and postoperative variables—affects major morbidity and mortality outcomes for LTx patients.

The COVID-19 Pandemic

The COVID-19 (coronavirus disease 2019) pandemic had a profound impact across health care delivery systems worldwide, and certainly deserves mention in this year’s review of LTx literature. Both direct and secondary consequences of this outbreak have affected patients along every stage of the LTx process. For example, while clear data are not yet available, it is possible that waitlist mortality was higher than normal because these patients have less reserve to tolerate infection with COVID-19, and in many regions, LTx volumes appear to have been lower than in years past.^13,14 The latter effect was driven by the obvious concerns over the risk of LTx recipients contracting COVID-19 postoperatively, but also due to constraints imposed by resource utilization factors—after all, LTx recipients require significant postoperative critical care and other inpatient, multidisciplinary therapies. In response, the COVID-19 pandemic has propelled the development of specific approaches to COVID-19 for perioperative LTx management that balance concerns regarding waitlist mortality with the goal of maintaining successful outcomes. ¹⁵

Another novelty arising from the pandemic was that end-stage lung disease secondary to COVID-19 became a new indication for LTx with unique considerations, ¹⁶ and several successful cases of LTx for COVID-19 were reported.^17-19 In addition, a graft from a donor who had previously recovered from infection with SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) infection was successfully implanted in a SARS-CoV-2 naïve recipient. ²⁰ Here, there were no major complications or other evidence of viral activation with immunosuppression, even though lung biopsy at the time of transplant was still positive for SARS-CoV-2 RNA by PCR (polymerase chain reaction) testing.

Adult Congenital Heart Disease (ACHD)

The number of congenital heart disease (CHD) patients surviving until adulthood has increased significantly over recent decades. In fact, the number of adults with CHD is now greater than the number of children with CHD. Traditionally, ACHD patients receiving a HTx have done poorly in the short term but have had equal to or better long-term outcomes than their non-ACHD counterparts. Single ventricle physiology with a failing Fontan is the most commonly transplanted physiology. The most commonly transplanted lesions are transposition of the great arteries and tricuspid atresia. ²¹

In 2020, there were multiple articles published on the survival of these patients following HTx. Becher et al ²² looked at all the ACHD HTxs from 1999 to 2015 in the Eurotransplant region. Two hundred and four ACHD patients were listed for heart, heart-lung, or heart-another organ transplantation during that time. Thirty-seven percent of the listed patients were on the “high urgency” list. Seventy-five percent of patients received a HTx with the others either dying while on the waitlist or being delisted because of a change in condition. Patients not included in this study were those who needed repeat HTx, and those with arrhythmogenic right ventricular cardiomyopathy or noncompaction. One hundred and five patients received heart alone, 76 received heart-lung, and 23 received heart plus another organ. Survival at 1 year was 73.2% and 66.6% at 5 years. Those receiving hearts alone had the best survival, with heart-lung next best and heart-another organ having the worst outcomes. Respiratory failure requiring mechanical ventilation and need for antiarrhythmics while listed were predictors of poor outcomes. In this cohort, failure of a systemic right ventricle was the number one indication for transplant.

Kinsella et al ²³ published outcomes of HTxs in the ACHD population versus matched controls. This was a single-center, retrospective review of all HTxs between 1998 and 2017. Three hundred and three HTxs were performed during that period either secondary to ACHD (n = 38), ischemic cardiomyopathy (ICM; n = 110) or dilated cardiomyopathy (DCM; n = 155). Regarding population characteristics, in the overall ACHD cohort there were more women than in the ICM or DCM cohorts. ACHD patients were also younger, more likely to need repeat sternotomy, have a lower body mass index, lower body surface area, and lower predicted heart mass. ACHD patients were also ranked lower on the acuity scale for wait listing. When the ACHD cohort was matched with controls, none of the above-mentioned differences were significant. Not surprisingly, cardiopulmonary bypass time was longer for the ACHD group. Mortality was not statistically different between the ACHD cohort and matched controls. Despite the lack of mortality difference, the authors did compute the relative risk of death and found that patients with ICM and DCM had a 66% lower chance of death compared with recipients with ACHD (P < .006).

Also this year, Riggs et al ²⁴ published an article designed to look at risk factors for 1-year mortality in ACHD patients receiving a HTx and how their outcomes compared with non-ACHD HTxs. The study included all HTx patients 17 years of age or older in the UNOS database between 2000 and 2018. The authors then divided the cohort into early (2000-2008) and late era (2009-2018) recipients. Late era recipients had better 1-year and long-term survival compared with the early era recipients. Risk factors for poor survival outcomes in ACHD patients compared with non-ACHD patients included body mass index greater than 25 kg/m², total bilirubin greater than 1.2 mg/dL, and glomerular filtration rate less than 60 mL/min/1.73 m². When late era ACHD patients had only 1 of these 3 mortality risk factors, their 1-year survival was the same as patients without ACHD. Late era ACHD patients with normal renal and hepatic function preoperatively had better long-term outcomes (74% survival at 7 years) compared with non-ACHD patients (71% survival at 7 years).

Patients with ACHD represent a physiologically diverse population with unique anesthetic considerations for thoracic transplant anesthesiologists. In 2020, Yuki ²¹ published an excellent review of the anesthetic management for HTx in the ACHD population. The work includes a thorough discussion of specific preoperative, intraoperative, and postoperative anesthetic challenges for these patients, as well as a review of the epidemiology and outcomes for ACHD patients undergoing HTx.

Donation After Circulatory Death (DCD)

We are all aware that the need for organs is greater than the supply. Optimizing grafts from DCD donors represents a potential avenue to expand the organ pool. Guariento et al ²⁵ explored the implantation of mitochondria into heart grafts after reperfusion. In their porcine model, circulatory death was induced in Yorkshire pigs, and after 20 minutes of warm ischemia the hearts were given cardioplegia and then placed on a perfusion system. There were 3 different arms from here. One arm received placebo, one received mitochondria, and one received mitochondria times 2 doses separated by 2 hours of perfusion and capped with 2 hours of perfusion after the second dose. There was a fourth group that was a sham for ischemic time. All hearts received 4 hours of reperfusion under a combination of loaded and unloaded conditions. Hearts that did have warm ischemic time but did not receive mitochondria had evidence of injury and edema that the mitochondria fed hearts did not have. These authors were able to show improved myocardial function and oxygen consumption with transplantation of mitochondrial cells into reperfused porcine hearts after 20 minutes of warm ischemic time.

Ngai et al ²⁶ published their experience with 4 DCD recipients in New York from June 2020 to March 2020. Donor age ranged from 26 to 44 years of age. Functional warm ischemic time (defined as systolic blood pressure less than 80 mm Hg to reperfusion) ranged from 27 to 35 minutes. All donors were managed with normothermic perfusion. Criteria for acceptance of the heart were mean arterial pressure greater than 60 mm Hg, central venous pressure less than 12 mm Hg, pulmonary artery systolic pressure less than 40 mm Hg, pulmonary capillary wedge pressure less than 12 mm Hg, myocardial volume oxygen greater than 65%, CI greater than 2.2 L/min/m², and transesophageal echocardiography demonstrating left ventricular ejection fraction greater than 50%, normal right ventricle systolic function, normal biventricular chamber size without wall motion abnormalities, E′ on tissue Doppler of 10 cm/s or greater and no valvular disease worse than mild. There was no maximum vasopressor dose to achieve these parameters, but the maximum inotrope was dobutamine at 5 µg/kg/min. All hearts were Maastricht category III. All recipients were extubated on postoperative day 1 and no one required mechanical circulatory support. Mean time to hospital discharge posttransplant was 12.5 days.

Messer et al ²⁷ published their experience with 79 DCD HTxs in Cambridge, United Kingdom. This was a retrospective, matched, observation cohort of HTxs from March 2015 through February 2020. All DCD donors were Maastricht category III. Primary endpoints were 30-day and 1-year survival. DCD hearts were harvested with 1 of 2 techniques. The initial technique was thoracoabdominal normothermic regional perfusion (TA-RP). Five minutes after cardiac death a sternotomy was performed, heparin was given, the right atrium and aorta were cannulated, and the arch vessels were stapled off. After exclusion of the arch vessels cardiac perfusion was restored by TA-RP. After cardiac function was normalized, TA-RP was weaned off, the heart was flushed with cold cardioplegia, procured, and placed on an organ care system (OCS). The other harvest system used, called Papworth DPP, involved similar steps except after heparinization the donor was exsanguinated and the blood was used to prime the OCS. Cold cardioplegia was then delivered to the heart, harvested, and then placed on the OCS system. All recipients were at Royal Papworth Hospital, Cambridge, UK. Inclusion criteria for recipients include a PVR less than 3 Woods units, transpulmonary gradient less than 12 mm Hg, and otherwise eligible for listing. Median functional warm ischemic time (systolic blood pressure less than 50 mm Hg to reperfusion by OCS) for donors was 24 minutes. There was no difference in 3-month mortality, 1-year mortality, or need for mechanical support between DCD and matched DBD recipients. Between harvest techniques there was no difference in mortality. There was a statistically significant increase in mechanical ventilation time (1.4 vs 0.5 days), hospital length of stay (24 vs 18 days), and need for hemofiltration (21 vs 2 patients) in the DPP group compared with the TA-RP group.

Total Artificial Heart (TAH) as a Bridge to Transplant

There were 2 studies in 2020 looking at outcomes after transitioning from the TAH to HTx. The first study we will explore is a single-center retrospective project out of France. This team’s goal was to look at the immediate and long-term outcomes of 50 patients from 1988 to 2019 who bridged from TAH-t to HTx. The operations were technically difficult and multiple chest reentry events occurred, including damage to the pneumatic line, aortic graft, and right ventricle. Other issues encountered were severe vasoplegia during explant in patients with mediastinitis, pulmonary edema in 1 patient whose device was turned off too early, and pericardial thickening and retraction requiring pericardiectomy in 21 of the 50 patients. Postoperatively, acute kidney injury (AKI) was seen in 72% of patients, with 42% of them requiring hemodialysis. Extracorporeal Life Support (ECLS) was required in 18% of the recipients and 22% were taken back to the operating room for bleeding. Despite these issues, survival was 74.8% at 4 months. Survival at 10 and 12 years were 57.9% and 53.5%, respectively. Most deaths were in the first year. Outcomes were best if patients were transplanted within 3 to 6 months of receiving the TAH-t. The authors point out 2 things that vary significantly from HTx in other patients. First, pericardial symphysis is a known problem after TAH-t implantation. Constriction of the new graft should be given significant consideration when hemodynamic instability is seen postoperatively. Second, chest reentry can cause issues we are not used to and superficial positioning of the driveline can cause significant problems. The authors recommend performing sternotomy on femoral-femoral cardiopulmonary bypass with the TAH-t flowing at 40 beats per minute. ²⁸

Carrier et al ²⁹ published their multicenter study on HTx after TAH-t in 6 high-volume centers. High-volume centers were defined as 10 or more TAH implantations a year. Patients were collected from 2014 to 2019. End points were survival and adverse events after TAH-t and HTx. Of those patients progressing to transplant after TAH-t, 55% received a HTx within 6 months and 76% received a HTx within 12 months of receiving their TAH-t. The most common adverse events after transplant were infection (46%), acute kidney injury (30%), respiratory failure (26.8%), neurologic dysfunction (21.7%), major bleeding (20%), and unplanned readmission (50%). Again, despite the significant occurrence of complications, survival was 84% at 1 year, 77% at 3 years, and 75% at 5 years. Age at TAH-t implantation was the most important predictor of death after HTx.

MitraBridge

In 2020, the International MitraBridge registry published its 1-year outcomes. MitraBridge is an international registry with 17 contributing centers from Europe and Canada that enrolls patients who are potential HTx candidates with (1) severe left ventricular dysfunction and/or NYHA (New York Heart Association) class III or IV symptoms and (2) moderate to severe or severe functional mitral regurgitation. All of the patients in this registry who received a MitraClip were either active on the HTx list, were bridge to decision or bridge to candidacy. The study goal was to record the 1-year composite adverse events of death from any cause, urgent HTx or left ventricular assist device placement, and first rehospitalization for heart failure. No deaths occurred during MitraClip placement or 30 days postprocedure. Fifteen percent of clipped patients received a HTx, and 23.5% no longer needed a HTx secondary to clinical improvements. For patients no longer needing HTx, 89% of these patients had a downgrade of NYHA class by 1 or more and none of them had NYHA class IV symptoms after clipping. Two thirds of the enrolled patients avoided death, urgent HTx, urgent ventricular assist device, and rehospitalization for heart failure. The authors believe these results show the MitraClip to be safe and effective in this patient population, thereby potentially expanding the MitraClip’s use in heart failure patients. ³⁰