Ann Thorac Cardiovasc Surg 2018; 24: Online January 26, 2018 doi: /atcs.oa Original Article

Transcription

1 Ann Thorac Cardiovasc Surg 2018; 24: Online January 26, 2018 doi: /atcs.oa Original Article Selective Cerebral Perfusion with the Open Proximal Technique during Descending Thoracic or Thoracoabdominal Aortic Repair: An Option of Choice to Reduce Neurologic Complications Katsuhiro Hosoyama, MD, PhD, Shunsuke Kawamoto, MD, PhD, Kiichiro Kumagai, MD, PhD, Masatoshi Akiyama, MD, PhD, Osamu Adachi, MD, PhD, Satoshi Kawatsu, MD, PhD, and Yoshikatsu Saiki, MD, PhD Purpose: Selective cerebral perfusion with the open proximal technique for thoracoabdominal aortic repair has not been conclusively validated because of its procedural complexity and unreliability. We report the clinical outcomes, particularly the cerebroneurological complications, of an open proximal procedure using selective cerebral perfusion. Methods: A retrospective chart review identified 30 patients between 2007 and 2015 who underwent aortic repair through left lateral thoracotomy with selective cerebral perfusion, established through endoluminal brachiocephalic and left carotid artery and retrograde left axillary artery. Results: The mean durations of the open proximal procedure and cerebral ischemia (the duration of the open proximal procedure minus the duration of selective cerebral perfusion) were ± 40.1 min and 24.8 ± 13.0 min, respectively. There were two cases (7%) of permanent neurologic dysfunction (PND) but no in-hospital deaths. Multivariate analysis identified the duration of cerebral ischemia as an independent risk factor for neurologic complications including temporary neurologic dysfunction (TND; odds ratio (OR): 1.13; p = 0.007), but no correlation was found between selective cerebral perfusion duration and neurologic complications. Conclusion: Despite the relatively long duration of the open proximal procedure, selective cerebral perfusion has a potential to protect against cerebral complications during thoracic aortic repair through a left lateral thoracotomy. Keywords: aortic repair, open proximal procedure, selective cerebral perfusion, circulatory arrest, neuroprotection Division of Cardiovascular Surgery, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan Received: August 8, 2017; Accepted: December 18, 2017 Corresponding author: Yoshikatsu Saiki, MD, PhD. Division of Cardiovascular Surgery, Tohoku University Graduate School of Medicine, 1-1 Seiryocho, Aoba-ku, Sendai, Miyagi , Japan yoshisaiki@med.tohoku.ac.jp 2018 The Editorial Committee of Annals of Thoracic and Cardiovascular Surgery. All rights reserved. Introduction Several operative strategies are available for open surgical repair of the descending thoracic aorta (DTA) or thoracoabdominal aorta (TAA), including the clamp-andsew technique, left heart bypass, partial cardiopulmonary bypass (CPB), and hypothermic CPB with circulatory arrest (CA). Surgical mortality is approximately 15%, Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018) 89

2 Hosoyama K, et al. and there are some complications, including stroke, which occurs in 3% 8% of cases. 1) Hypothermic CPB with intervals of CA is increasingly being used. Its advantages include its excellent neuroprotective effect, elimination of the need for proximal clamping, and a dry, bloodless field. However, cerebral ischemia is a disadvantage of this approach. Generally, the permissible duration of CA is min in a deep hypothermic condition. 2 5) However, a longer duration of CA can sometimes be needed. For example, redo surgery after thoracic endovascular aortic repair, infectious aneurysm, and cases with Marfan syndrome require meticulous dissection and complex proximal anastomosis. In these situations, there is an increased risk of cerebral complications because of the longer duration of cerebral ischemia. The use of selective cerebral perfusion (SCP) for DTA or TAA repair through a left lateral approach could be an option to address this risk; however, this has not yet been conclusively validated because of its procedural complexity and potential unreliability. Since 2007, we have employed a deep hypothermic open proximal procedure with SCP for DTA and TAA repair through a left thoracotomy in selected cases to avoid neurologic complications and unwanted retrograde aortic dissection. The aim of this study was to assess the clinical outcomes, particularly cerebroneurological complications, of this open proximal procedure with SCP. Materials and Methods Subjects We conducted a retrospective chart review of 30 patients who underwent DTA or TAA repair using deep hypothermic CA with SCP in the Division of Cardiothoracic Surgery at Tohoku University Hospital between January 2007 and December All patients provided written informed consent, and the ethics committee and the institutional review board of our institution approved the study protocol (application number ). Operative technique The right atrium (via the femoral vein) and the pulmonary artery were cannulated for venous drainage. The arterial inflow came from the femoral and left axillary arteries, and left atrial venting was also performed. For the SCP circuit, the brachiocephalic artery (BCA) and left common carotid artery (CCA) were cannulated in addition to the left axillary arterial cannulation (Fig. 1A). The left lateral thoracotomy approach was performed through the fifth or sixth intercostal space. To secure an adequate visualization of the distal aortic arch, the fourth and fifth ribs were transected posteriorly, if needed. When the bladder temperature reached 20 C, the distal aortic arch was opened. After cardiac standstill was achieved with cold blood cardioplegia in an antegrade manner using an 18-French Foley balloon catheter, the BCA and left CCA were cannulated endoluminally using SP Stud catheters with a nonslip silicone lattice (16 French for the BCA and 12 French for the left CCA; Fuji Systems Co., Tokyo, Japan), and the proximal part of the left subclavian artery (SCA) was secured with a tourniquet (Fig. 1B). SCP was commenced with two roller pumps, each at an initial pump flow rate of 5 ml/kg/min; subsequently, the flow was adjusted according to the regional oxygen saturation at the bilateral hemispheres of the brain. During the open proximal procedure, the distal side of the aorta was perfused via femoral arterial inflow. The cerebral perfusion was established through the BCA, left CCA, and left SCA in 26 (87%) patients and through the BCA and left CCA in four (13%) patients. For the proximal anastomosis, we used a modified cuffed anastomosis technique, as previously described. 6) In brief, several U-stay sutures were placed between the graft and the aorta, the graft was rolled back approximately 5 mm, and the inverted graft was placed into the open end of the aorta. A continuous running suture was then placed circumferentially, with felt strip reinforcement. Statistical analysis Categorical variables are presented as frequencies and percentages, and continuous variables as mean ± standard deviation. Statistical significance was determined with the χ 2 test or Fisher s exact test for categorical variables and Student s t-test or the Mann Whitney U-test for continuous variables. All reported p-values were twosided, and p < 0.05 was considered statistically significant. Pre- and intraoperative factors were further analyzed using a multivariate logistic regression model to identify risk factors for neurologic complications. Parameters that showed a significant effect on neurologic complications in the univariate analysis were included in the multivariate model. The results are expressed as odds ratios (ORs) with 95% confidence intervals (CIs). All statistical analyses were performed using JMP version 11.0 (SAS Institute Inc., Cary, NC, USA). Permanent neurologic dysfunction (PND) was defined as the presence of focal (stroke) or global (coma) injury 90 Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018)

3 Selective Cerebral Perfusion via Left Thoracotomy A FA cannulation RA drainage SCP (BCA) (via FV) PA cannulation SCP (left CCA) LA venting left AxA rso 2 B Fig. 1 (A) The extracorporeal circuit and SCP circuit. (B) A schematic representation of SCP. AxA: axillary artery; BCA: brachiocephalic artery; CCA: common carotid artery; FA: femoral artery; FV: femoral vein; LA: left atrium; PA: pulmonary artery; RA: right atrium; rso 2 : regional saturation of oxygen; SCP: selective cerebral perfusion that did not resolve by the time of discharge. Temporary neurologic dysfunction (TND) was defined as symptoms of prolonged postoperative obtundation, disorientation, and persistent loss of cognitive function with no correlate on the computed tomography (CT) scan, with complete resolution before discharge. Results Baseline characteristics and intraoperative profiles The baseline characteristics of the patients are shown in Table 1. Chronic dissecting aneurysm (n = 22, 73%) was the most common pathology. Six patients (20%) had Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018) 91

4 Hosoyama K, et al. A B Freedom from All-cause Death Freedom from Aortic Event Years Fig. 2 (A) Kaplan Meier curve for cumulative survival to 8 years. (B) Kaplan Meier curve for freedom from aortic event to 8 years. Years Table 1 Baseline characteristics of patients undergoing descending thoracic aorta or thoracoabdominal aorta repair Baseline characteristics N = 30 Age, years 58.2 ± 13.2 Female, n (%) 9 (30) Hypertension, n (%) 25 (83) Dyslipidemia, n (%) 9 (30) Diabetes mellitus, n (%) 4 (13) Chronic obstructive pulmonary disease, n (%) 5 (17) Renal impairment (creatinine 1.5 mg/dl), n (%) 3 (10) History of cerebrovascular disease, n (%) 3 (10) Aortic pathology, n (%) Dissecting aneurysm 22 (73) Infectious cases 5 (2) Degenerative aneurysm 2 (7) Aortic invasion of lung cancer 1 (3) Marfan syndrome, n (%) 6 (20) Ruptured cases, n (%) 2 (7) Previous aortic repair, n (%) 11 (37) Left ventricular ejection fraction, % 66.0 ± 7.9 Marfan syndrome and 11 (37%) had a history of aortic surgery. With regard to Japan score, the average of 30 days operative mortality was 7.3% ± 6.3% and that of 30 days mortality and morbidity was 27.6% ± 17.6%. The operative data are shown in Table 2. The mean durations of the CPB and the open proximal procedure were ± min and ± 40.1 min, respectively. The duration of cerebral ischemia, that is, the time with no cerebral perfusion, calculated as the duration of the open proximal procedure minus the duration of SCP, was 24.8 ± 13.0 min. Deep hypothermic CA with SCP was selected for these patients because aortic clamping distal to the left CCA or the left SCA was considered inappropriate, either because of the lack of sufficient space (in 16 patients [53%]) or because of the poor pathological condition, such as extensive atheroma or calcification of the clamp site (in 14 patients [47%]). For two patients, the intraoperative measurement of cerebral regional oxygen saturations using near-infrared spectroscopy showed transient desaturation of more than 20% from baseline during SCP, which is indicative of stroke with a high sensitivity and specificity; 7) however, this was promptly corrected in both cases by appropriate adjustment of the SCP cannulae. Early postoperative outcomes The early postoperative outcomes are summarized in Table 2. There was no case of in-hospital death. PND occurred in two patients (7%). One of these patients suffered a left cerebellar infarction without any sequelae; 92 Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018)

5 Table 2 Operative data and early outcomes Selective Cerebral Perfusion via Left Thoracotomy N = 30 Emergent/urgent, n (%) 7 (23) Extent of repair, n (%) Descending aorta 11 (37) Thoracoabdominal aorta: Crawford type 1 8 (27) Crawford type 2 11 (37) CSF drainage, n (%) 20 (67) CPB duration, min ± Open proximal duration, min ± 40.1 SCP duration, min 85.4 ± 40.3 Cerebral ischemia duration, min 24.8 ± 13.0 Lowest body temperature, C 20.3 ± 1.5 Minimum cerebral rso 2 during SCP Right side, % 67.2 ± 11.5 Left side, % 66.0 ± 12.6 In-hospital death, n 0 PND, n (%) 2 (6.7) TND, n (%) 11 (36.7) Paraplegia, n 0 Bleeding re-exploration, n (%) 3 (10) Low cardiac output syndrome, n (%) 3 (10) Arrhythmic event, n (%) 7 (23) Requirement for new dialysis, n (%) 3 (10) Prolonged ventilation (>24 h), n (%) 26 (86.7) Tracheostomy, n (%) 8 (26.7) CSF: cerebrospinal fluid; CPB: cardiopulmonary bypass; SCP: selective cerebral perfusion; rso 2 : regional oxygen saturation; PND: permanent neurologic dysfunction; TND: temporary neurologic dysfunction in this case, a certain embolic cause was suspected rather than a hypoperfusional mechanism. The other patient suffered subarachnoid hemorrhage and right hemiparesis, possibly resulting from overdrainage of cerebrospinal fluid. TND occurred in 11 patients (37%), all of whom fully recovered before discharge. Of these 11 cases of TND, 2 had delayed awakening and the others had persistent agitated condition. The average of the cerebral ischemic time was 28.5 ± 23.3 min in patients with PND (n = 2) and 33.6 ± 12.7 min in patients with TND (n = 11). The two patients described earlier who exhibited transient regional oxygen desaturation during suboptimal SCP did not develop any neurologic complications postoperatively. Analysis of risk factors for neurologic complications In the univariate analysis, the preoperative factors of age (p = 0.01) and chronic obstructive pulmonary disease (p = 0.07), as well as the intraoperative factor of cerebral ischemic duration (p = 0.001), were identified as possible risk factors for postoperative neurologic complications, including both PND and TND (Table 3). The multivariate analysis identified age (OR: 1.08, 95% CI: , p = 0.047) and the duration of cerebral ischemia (OR: 1.13, 95% CI: , p = 0.007) as risk factors for neurologic complications (Table 3). Late outcomes Follow-up was 100% completed. The average followup period was 2.7 ± 2.0 years (range: years). Overall survival at 2, 4, and 6 years was 91%, 78%, and 78%, respectively (Fig. 2). There were three cases of late death: one patient died of pneumonia, one of sepsis after a digestive surgery, and one of congestive heart failure 4 years after the aortic surgery. The aortic eventfree rate at 2, 4, and 6 years was 90%, 83%, and 83%, respectively. Four patients suffered aorta-related complications. Of these, two underwent additional radiologic intervention for abdominal malperfusion and pseudoaneurysm, respectively, one underwent scheduled surgery for a visceral artery aneurysm, and one underwent redo surgery due to dilatation of the remaining native aorta. Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018) 93

6 Hosoyama K, et al. Table 3 Variable Univariate and multivariate analyses of pre- and intraoperative risk factors for neurologic complications Univariate analysis Multivariate analysis OR (95% CI) p value OR (95% CI) p value Age 1.08 ( ) ( ) Female 2.03 ( ) 0.38 Hypertension 1.18 ( ) 0.87 Dyslipidemia 2.03 ( ) 0.38 Diabetes mellitus 0.39 ( ) 0.41 Chronic obstructive pulmonary disease 7.11 ( ) ( ) 0.77 Renal impairment (creatinine 1.5 mg/dl) 0.71 ( ) 0.71 History of cerebrovascular disease 0.63 ( ) 0.71 Pathology: dissecting 1.39 ( ) 0.70 Marfan syndrome 1.40 ( ) 0.71 Previous aortic repair 0.63 ( ) 0.56 Left ventricular ejection fraction 0.94 ( ) 0.20 Emergent/urgent 0.98 ( ) 0.98 Extent of repair: DTA 1.15 ( ) 0.86 Extent of repair: Crawford type ( ) 0.20 Extent of repair: Crawford type ( ) 0.17 CSF drainage 2.33 ( ) 0.29 SCP duration 1.01 ( ) 0.18 Cerebral ischemia duration 1.13 ( ) ( ) OR: odds ratio; CI: confidence interval; DTA: descending thoracic aorta; CSF: cerebrospinal fluid; SCP: selective cerebral perfusion Discussion Despite continuing improvements in neuroprotective strategies, postoperative cerebroneurological complications following DTA or TAA surgery still occur in 3% 8% cases. 1) In particular, the risk of cerebral complications increases when the distal aortic arch is not suitable for proximal clamping, as is the case for some Crawford type 1 or 2 aneurysms, for which it is necessary to perform an anastomosis of the proximal aorta with the use of CA. As yet, it has not been clarified how cerebral hypoperfusion affects postoperative neurologic complications following DTA or TAA surgery. However, Yoo et al. found that the incidence of stroke did not differ significantly between groups of patients treated with anastomosis using CA and with distal arch clamping. 8) This suggests that embolic events, such as the formation of atheromatous plaques or thrombi, are not the only cause of postoperative neurologic complications. We believe that cerebral hypoperfusion during CA may be another factor that contributes to neurologic complications. Ergin et al. reported that longer CA increased the incidence of TND following aortic arch repair through a median approach. 4) We therefore postulated that the use of SCP for DTA or TAA surgery may help in reducing cerebroneurological complications. Since 2007, we have used the SCP technique for DTA and TAA surgery for which we expected a substantially lengthy duration of CA, such as redo surgery after thoracic endovascular aortic repair and cases of infectious aneurysm or Marfan syndrome. The advantage of SCP is that it permits a longer duration of CA, allowing more complex aortic repairs to be performed. Okada et al. 9) and Hino et al. 10) reported that SCP provided excellent brain protection during aortic repair through a left thoracotomy; however, both of their patient cohorts underwent the one-stage repair of extended thoracic aneurysms and the approach to the aortic arch differed from our technique. Hagl et al. 11) noted that adjunctive SCP increased the permissible duration of CA from 40 min to 80 min for aortic arch surgery via a median sternotomy. Similarly, Eusanio et al. 12) found that a CA duration of greater than 25 min was associated with an increased risk of TND in ascending aortic repair for acute type A dissection, whereas the duration of SCP had no effect on the neurologic outcome. Kawaharada et al. reported the incidence rate of cerebral embolism as 3.4% in DTA or TAA repair under deep hypothermic CA for a mean duration of 28 min, and advocated the use of right axillary arterial perfusion. 13) In comparison, our patient cohort had a higher incidence of PND (6.7%). However, it is noteworthy that the mean duration of SCP for our 94 Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018)

7 Selective Cerebral Perfusion via Left Thoracotomy cohort was 85.4 min, with a mean duration of cerebral ischemia of 24.8 min, implying that the reparative procedures undertaken in our series were of much longer duration. Indeed, 73% of patients in our cohort underwent chronic aortic dissection in which complete obliteration of the false lumen was achieved through modified cuffed anastomosis, as described earlier, which was time-consuming. Considering the duration of the procedure was so long, the incidence of PND might be considered acceptable. However, a disadvantage of using SCP during left lateral thoracotomy is its complexity and the possibility of the accidental embolization of atheromatous debris during manipulation of the arch vessels. Nevertheless, we believe that these issues can be addressed with methodical care during cannulation, careful de-airing, and, in some cases, the use of retrograde cerebral perfusion at the termination of CA to flush out emboli. In this study, we defined neurologic complications as PND and TND. Ergin et al. 14) found that TND was correlated with poor neurologic function at 6 weeks postoperatively. In other words, TND may not be a benign self-limiting condition but may be related to significant neurologic injury. Previous studies have identified risk factors for TND as being preoperative cerebrovascular disease, patient age, undergoing an emergent operation, and a long duration of CA. 4,15 18) The present study showed that limiting the duration of cerebral ischemia could reduce the incidence of neurologic complications. The use of SCP for DTA or TAA surgery, resulting in a shorter period of cerebral ischemia, may reduce the risk of TND. In this cohort, the postoperative pulmonary dysfunction was observed relatively frequently; however, it is not necessarily attributable to SCP technique. Postoperative prolonged mechanical ventilation is also possibly linked with considerably high preoperative risk profile in this cohort. It was eventually resolved and did not affect early and long-term morbidity or mortality. Several limitations of the present study warrant consideration. This investigation was subject to the limitations inherent in retrospective single-arm studies that use observational data. It represents the 8-year experience of a single, large tertiary referral center, and the results may not be generalizable to other centers. Patient outcomes were documented by direct examination or through communication with family members or referring physicians. The number of patients included in this study was relatively small. Further studies in a larger population are needed to obtain more definitive results. Conclusion Despite the relatively long duration of the open proximal procedure, the use of SCP for DTA or TAA repair via the left lateral thoracotomy approach possibly affords a protective effect against cerebral complications. The results of multivariate analysis indicated that the use of SCP for left lateral aortic surgery may reduce the occurrence of neurologic complications by limiting the duration of cerebral ischemia. Further studies are needed to clarify the clinical characteristics of cohorts in which the use of SCP is particularly effective for reducing the risk of neurologic complications. Disclosure Statement The authors have declared that no conflicts of interest exist. References 1) Piazza M, Ricotta JJ. Open surgical repair of thoracoabdominal aortic aneurysms. Ann Vasc Surg 2012; 26: ) McCullough JN, Zhang N, Reich DL, et al. Cerebral metabolic suppression during hypothermic circulatory arrest in humans. Ann Thorac Surg 1999; 67: ; discussion ) Svensson LG, Crawford ES, Hess KR, et al. Deep hypothermia with circulatory arrest. Determinants of stroke and early mortality in 656 patients. J Thorac Cardiovasc Surg 1993; 106: 19-28; discussion ) Ergin MA, Galla JD, Lansman sl, et al. Hypothermic circulatory arrest in operations on the thoracic aorta. Determinants of operative mortality and neurologic outcome. J Thorac Cardiovasc Surg 1994; 107: ; discussion ) Kayatta MO, Chen EP. Optimal temperature management in aortic arch operations. Gen Thorac Cardiovasc Surg 2016; 64: ) Matsuo S, Oda K, Motoyoshi N, et al. Modified cuffed anastomosis technique to treat pseudoaneurysms following thoracic endovascular aortic repair. Interact Cardiovasc Thorac Surg 2012; 14: ) Olsson C, Thelin S. Regional cerebral saturation monitoring with near-infrared spectroscopy during selective antegrade cerebral perfusion: diagnostic performance and relationship to postoperative stroke. J Thorac Cardiovasc Surg 2006; 131: ) Yoo JS, Kim JB, Joo Y, et al. Deep hypothermic circulatory arrest versus non-deep hypothermic beating heart strategy in descending thoracic or thoracoabdominal aortic surgery. Eur J Cardiothorac Surg 2014; 46: Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018) 95

8 Hosoyama K, et al. 9) Okada K, Tanaka A, Munakata H, et al. Extended replacement of aortic arch aneurysms through left posterolateral thoracotomy. Eur J Cardiothorac Surg 2009; 35: ) Hino Y, Okada K, Oka T, et al. Extended replacement of the thoracic aorta. Eur J Cardiothorac Surg 2013; 43: ; discussion ) Hagl C, Ergin MA, Galla JD, et al. Neurologic outcome after ascending aorta-aortic arch operations: effect of brain protection technique in highrisk patients. J Thorac Cardiovasc Surg 2001; 121: ) Di Eusanio M, Wesselink RM, Morshuis WJ, et al. Deep hypothermic circulatory arrest and antegrade selective cerebral perfusion during ascending aorta-hemiarch replacement: a retrospective comparative study. J Thorac Cardiovasc Surg 2003; 125: ) Kawaharada N, Ito T, Naraoka S, et al. Right axillary arterial perfusion for descending thoracic or thoracoabdominal aortic aneurysm repair with open proximal anastomosis through left thoracotomy. Gen Thorac Cardiovasc Surg 2014; 62: ) Ergin MA, Uysal S, Reich DL, et al. Temporary neurological dysfunction after deep hypothermic circulatory arrest: a clinical marker of long-term functional deficit. Ann Thorac Surg 1999; 67: ; discussion ) Di Eusanio M, Schepens MA, Morshuis WJ, et al. Antegrade selective cerebral perfusion during operations on the thoracic aorta: factors influencing survival and neurologic outcome in 413 patients. J Thorac Cardiovasc Surg 2002; 124: ) Kazui T, Yamashita K, Washiyama N, et al. Usefulness of antegrade selective cerebral perfusion during aortic arch operations. Ann Thorac Surg 2002; 74: S1806-9; discussion S ) Dossche KM, Morshuis WJ, Schepens MA, et al. Bilateral antegrade selective cerebral perfusion during surgery on the proximal thoracic aorta. Eur J Cardiothorac Surg 2000; 17: ) Di Bartolomeo R, Pacini D, Di Eusanio M, et al. Antegrade selective cerebral perfusion during operations on the thoracic aorta: our experience. Ann Thorac Surg 2000; 70: 10-5; discussion Ann Thorac Cardiovasc Surg Vol. 24, No. 2 (2018)