Discussion
The FET procedure is a promising and an evolving technique that is associated with acceptable rates of mortality and morbidity compared to the conventional elephant trunk procedure in the management of complex thoracic aortic pathologies.51 The development of coagulopathy (i.e., reoperation for bleeding and blood products transfusion) in aortic surgery is commonly observed in patients undergoing repair of aneurysm or dissection.52,53According to a position paper reported by the Vascular Domain of EACTS, postoperative re-exploration because of bleeding ranged between 2.5-30%.54 The association between CPB or HCA duration and perioperative coagulopathy risk is well-documented and reported in literature. Mazzeffi et al. highlight that DHCA duration and the need for RBC transfusion are closely associated, and that this is even more pronounced with CPB durations between 120 and 180 minutes.55 Similarly, decreased platelet count and function are well-documented sequelae of pediatric cardiac and aortic reconstruction involving CPB and DHCA.56, 57
No studies have explored the prevalence of coagulopathy specifically in patients undergoing arch replacement using the FET technique. In their meta-analysis, Tian et al. documented the safety and efficacy of the frozen elephant trunk technique and reported a pooled mortality of 8.3%, stroke rate of 4.9%, and spinal cord injury rate of 5.1%.58 Though the findings of Tian et al. are robust, our study demonstrated a strong association between the FET procedure for total arch replacement and perioperative coagulopathy, represented by reoperation for postoperative bleeding and the need for blood product transfusion. 17% of the patients included suffered postoperative bleeding, while the pooled estimate of reoperation for postoperative bleeding was 7%.
The mean volumes of transfused packed blood cells and fresh frozen plasma were 1677 ml and 1016 ml respectively.
The pathophysiology of surgery-induced coagulopathy in aortic operations has not been properly described; however, it has been reported that acute aortic dissection itself activates hemostatic systems before surgery.23 Liu et al59 evaluated perioperative consumption coagulopathy among sixty-six patients with acute type A aortic dissection against thirty-six patients with thoracic aortic aneurysms. They demonstrated that clotting factors and fibrinogen levels were perioperatively reduced, but platelet functions were less changed. Nomura et al60assessed the coagulation states after type A aortic dissection surgery and followed patients up for 82 months. In their study, hypercoagulability was identified even in the chronic phase, as measured by D-dimer and thrombin–antithrombin III complex. On the other hand, Li et al61 studied patients with aortic dissection undergoing surgical repair and found that the relationship between the coagulation pathway and the postoperative inflammatory reaction should be considered in future studies. In this review, we were not able to explore any changes in blood parameters and inflammatory markers perioperatively, but we found a substantial need for blood product transfusion after surgery, suggesting the paramount effects of consumption coagulopathy on patients undergoing aortic surgeries. Further investigations about the coagulopathy-related factors and inflammatory markers could elucidate the proper application of such tests in the evaluation of patients undergoing the FET procedure for aortopathies.
The volume-outcome relationship associated with FET procedures is also worthy of discussion in the context of coagulopathy risk following FET. Mori et al. reported that lower-volume centres (fewer than 5 cases per year, reportedly making up 58.3% of centres) were at an increased risk of mortality (OR 2.50, 95% CI 2.08 - 3.01, p < 0.0001) compared to high-volume centres.62 This is unsurprising as greater patient volume usually translates to more opportunities to practice intricate interventions. It has not been conclusively demonstrated that increased operating volume in FET repair translates to shorter CPB/HCA durations, but would it not be reasonable to suggest that, faced with the same patient, the more experienced surgeon would be able to accomplish the same standard of repair in a shorter duration than a novice surgeon? This would allow rewarming to occur earlier, and possibly therefore reducing the risk of coagulopathy risk in FET. The effect of longitudinal surgical training in FET implantation and device selection on postoperative outcomes could also be investigated – indeed, much research is required to substantiate this.
Despite the abundance of literature detailing the hemodynamic sequelae associated with aortic repair procedures - ranging from continued bleeding to impaired platelet function - any clinical correlation between coagulopathies and individual FET device characteristics hitherto remains to be elucidated.63 In the present review we attempted to identify factors contributing to the development coagulopathy as well as sources of heterogeneity. Stent length, type, and surgical acuity were associated with a reduction in the heterogeneity of pooled estimate of reoperation for postoperative bleeding. In addition, in meta-regression analysis we found that both male and younger age were associated with a reduced risk of reoperation for postoperative bleeding.
No studies have evaluated predictors of coagulopathy among patients undergoing the FET procedure. Further investigations with a focus on finding predictors of bleeding and of coagulation events would be of great use to guide surgeons for risk stratification of patients scheduled for the FET procedure. For example, FET graft size and length is a well-known area of contention amongst aortic surgeons, and there currently exists no consensus around on the optimal approach to hybrid prosthesis sizing.63
Evidence on the effect of hybrid prosthesis sizing on clinical outcomes is varied – under-sizing has been reported as falling short of achieving an optimal seal, while over-sizing has been associated with further intimal injury and distal stent-induced entry tears with endoleak.64 Because suboptimal graft sizing can potentially lead to endoleak, perhaps necessitating reintervention to address postoperative bleeding, graft sizing is a pertinent area for investigation as a potential predictor or risk factor of FET-associate coagulopathy. Indeed, there exists robust data on the relationship between neurological outcome and FET graft sizing. For example, Preventza et al. report that stent lengths below 10 cm carried a lower risk of SCI, and Tan et al. suggest that proximalisation of TAR to Zone 0 may be associated with improved neurological outcomes.65, 66 Undoubtedly, a similarly nuanced appreciation of coagulopathy predictors associated with different FET devices and sizes would be of great benefit. This would facilitate a more holistic understanding of the risks and benefits associated with each individual procedure and allow a more nuanced evaluation of patient-specific needs and risks.
Apart from graft sizing, several FET devices are available,67-71 and their usage in clinical practice is evolving so that studies for outcomes of such devices are being increased in literature.51 The globally used tetrafurcate Thoraflex Hybrid®stent-graft provides separate reimplantation of supra-aortic vessels, an easier anastomosis to the distal aortic arch, and radio-opaque markers in the stented part.
In the subgroup analysis, we also found that Thoraflex® and Frozenix® had no heterogeneity compared to moderate heterogeneity depicted for the E-vita® and the Cronus® devices. However, it’s imperative to elicit that the wide generalization and distribution of Thoraflex Hybrid across diverse population groups globally and in line with different ethnic population applicability in concurrent to the streamed analysis in this review potentially raises the standards that Thoraflex Hybrid is considered as the gold standard. The global availability and wide application coupled with the multiplicity of the Thoraflex Hybrid marketing prowess and in alignment to clinical and surgical need with single perfusion branch and multi branch configuration have been utilized by many aortovascular surgeons who operate under diverse healthcare systems and have been apprenticed and trained differently. Henceforth, undoubtedly this could potentially explain the elicited heterogeneity of this device and its coagulopathy correlative outcomes as perceived comparatively to others FET including the Frozenix® and Cronus®.
Our analysis also revealed that the rate of reoperation for postoperative bleeding was the lowest the Thoraflex Hybrid® and for the Frozenix® among studies included in our proportional meta-analysis. Bearing these findings in mind it is worth highlighting that in a systematic review comparing the Thoraflex Hybrid® and E-vita® devices, Harky et al. emphasized that Thoraflex Hybrid® is more frequently implanted in cases of emergent aortic repair than E-vita®. Notwithstanding the effect of baseline patient demographics on clinical outcomes, this supports our finding that Thoraflex Hybrid® was associated with lower rates of coagulopathy than E-vita, with minimal heterogeneity.72 On the other hand, a recent report noted excessive oozing through the fabric of stent in three patients implanted the branched Cryolife-Jotec E-vita open NEO hybrid prosthesis.73 The E-vita open Neo HP has been described as having a propensity for excessive oozing through its polyester wall, which lacks gelatin or collagen (a key component of preventing graft porosity). This significant limitation has been associated with poor postoperative outlook, and Czerny et al. describe one such patient requiring 23 U of RBC, 28 U of FFP, and 16 U of thrombocytes in the 24 hours following surgery. The postoperative courses of the patients affected by this porosity were complicated by factors including low cardiac output necessitating inotropes, haemodialysis, mutli-organ failure, paraplegia, and death.73
Though Jakob et al. argue that strategic employment of ROTEM may help deal with the graft’s porosity, it is imperative to clinically point out that such an approach to FET-associated coagulopathy has been tested only on a small number of EVITA Neo, and no other FET devices.74
The heterogeneity of coagulopathy rates associated with the Cronus® and the E-vita® might be attributed to the diverse patient demographics (e.g., ethnicity) among studies reporting outcomes from E-vita® and because many such studies failed to report the major factors causing complications, rather than extracted variables in our meta-analysis.