Hitoshi Terada

Total arch replacement using aortic arch branched grafts with the aid of antegrade selective cerebral perfusion

BACKGROUND We report our clinical experience with total arch replacement using aortic arch branched graft in an attempt to determine the independent predictors of both in-hospital mortality and neurologic outcome. METHODS We studied 220 consecutive patients who underwent total arch replacement using aortic arch branched graft between May 1990 and June 1999. All operations were performed with the aid of hypothermic extracorporeal circulation, antegrade selective cerebral perfusion, and open distal anastomosis. RESULTS The overall in-hospital mortality rate was 12.7%. Multivariable analysis showed independent determinants of in-hospital mortality to be chronic renal failure, long pump time, participation in early series, and shock. Postoperative permanent neurologic dysfunction was 3.3%. On multivariable analysis, old cerebral infarct and pump time were independent determinants of permanent neurologic dysfunction. The selective cerebral perfusion time had no significant influence on in-hospital mortality or neurologic outcome. The 5-year survival rate including in-hospital deaths was 79% +/- 6%. CONCLUSIONS Selective cerebral perfusion allows increased ease of performance of total arch replacement, a complex and time-consuming procedure, and helps reduce periprocedural mortality and morbidity in patients with aortic arch aneurysm and those with acute aortic dissection.

Role of biologic glue repair of proximal aortic dissection in the development of early and midterm redissection of the aortic root

BACKGROUND Redissection of the aortic root after supracommissural aortic graft replacement with reapproximation of the layers of the dissected aortic root is relatively rare. Causes and surgical treatment of this lesion remain controversial. METHODS From January 1983 to September 2000, 130 patients had emergency operation for acute type A aortic dissection. Of them, 57 patients underwent root reconstruction using biologic glues and 4 patients (7.0%) developed redissection of the aortic root associated with moderate to severe aortic regurgitation 5 to 27 months after the initial operation. In all patients, the proximal false lumen was obliterated with infusion of gelatin-resorcinol-formaldehyde (GRF) glue or BioGlue and the aorta was reinforced with Teflon felt strip or Surgicel placed on its outside wall. RESULTS During reoperation, the noncoronary aortic sinus was found to be redissected in all patients with the dissection extending retrogradely to the aortic annulus. This resulted in aortic regurgitation with prolapse of the noncoronary cusp because the proximal suture line dehisced. Histopathology showed disappearance of the nuclei of the medial smooth muscle cells, suggesting tissue necrosis at the site of GRF glue application. The lesions were treated successfully with full root replacement using a freestyle heterograft bioprosthesis or a composite graft prosthesis. CONCLUSIONS The use of biologic glues for reapproximating the layers of the dissected aortic root is associated with a certain amount of risk of aortic wall necrosis. Therefore, care should be taken to ensure proper use of these glues. Full root replacement could be a preferable technique for treating redissection of the aortic root.

Extended total arch replacement for acute type a aortic dissection: Experience with seventy patients

OBJECTIVE We sought to report the clinical experience with extended total arch replacement for acute type A aortic dissection and to determine the factors that influence early mortality, late survival, and late reoperation. METHODS Between December 1988 and August 1998, 70 patients underwent emergency graft replacement of both the ascending aorta and the total aortic arch for acute type A aortic dissection. All operations were performed with hypothermic extracorporeal circulation, selective cerebral perfusion for cerebral protection during aortic arch repair, and open distal anastomosis. Concomitant procedures included aortic valve resuspension in 18 patients, composite graft replacement in 10 patients, and coronary artery bypass grafting in 5 patients. RESULTS The early mortality rate was 16% (11 of 70 patients). Multivariable analysis showed that renal-mesenteric ischemia and coronary artery bypass grafting were independent determinants for early death. Survival rates at 3 and 5 years postoperatively, including the early deaths, were 75% +/- 5% and 73% +/- 6%, respectively. Multivariable analysis showed that renal-mesenteric ischemia and en bloc repair were independent determinants for late death. Freedom from reoperation was 91% +/- 4% and 77% +/- 8% at 3 and 5 years, respectively. Multivariable analysis showed that anastomotic leakage was the only significant determinant for late reoperation. CONCLUSIONS Extended total arch replacement for acute type A aortic dissection could be justified in properly selected patients.

Usefulness of antegrade selective cerebral perfusion during aortic arch operations

BACKGROUND To evaluate the safety and usefulness of antegrade selective cerebral perfusion (SCP) during arch aneurysm or aortic dissection operations. METHODS Between January 1986 and December 2001, 330 patients underwent aortic arch repair using SCP. Operations were performed with the aid of hypothermic extracorporeal circulation, SCP, and systemic circulatory arrest in most cases. In all, 89 patients (27%) were operated on for acute aortic dissection, 77 (23%) for chronic aortic dissection, and 164 (50%) for degenerative aneurysm. Total arch replacement using a branched graft was performed in 288 patients (94%). Mean SCP time was 86.2 +/- 28.5 minutes. RESULTS The overall in-hospital mortality rate was 11.2% (falling to 3.2% in the 124 patients operated on between 1997 and 2001). Independent determinants of hospital mortality were pump time, renal/mesenteric ischemia, chronic renal failure, increasing age, period of operation, and nonuse of four-branched arch graft. The overall postoperative incidences of temporary and permanent neurologic dysfunction were 4.2% and 2.4%, respectively. There was no significant correlation between SCP time and in-hospital mortality or neurologic outcome. CONCLUSIONS Selective cerebral perfusion is a reliable technique for cerebral protection and it facilitates complex and time-consuming total arch replacement.

Impact of an aggressive surgical approach on surgical outcome in type a aortic dissection

BACKGROUND To evaluate the impact of an aggressive surgical approach on early and late outcome in type A aortic dissection. METHODS From 1983 to 2001, 240 patients underwent operation for acute (n = 138) and chronic (n = 102) type A aortic dissection. The extent of distal aortic resection included the ascending aorta in 39 (16%) patients, hemiarch (HAR) in 47 (20%), and total arch (TAR) in 154 (64%), including 19 patients who also had their descending aorta replaced (DAR). RESULTS The in-hospital mortality did not differ between TAR with or without DAR and other more conservative techniques (12.3% versus 16.3%). Actuarial survival at 10 years including in-hospital mortality was 72.4% +/- 3.3% and freedom from reoperation was 77.2% +/- 3.6% for all patients: neither was influenced by the extent of distal aortic resection or acuity of aortic dissection. Multivariate analysis revealed younger age and failure to resect the intimal tear to be independent determinants for late reoperation. However, in contrast to 22 patients who had more conservative operations, none of the patients with TAR required reoperation on the aortic arch through a sternotomy incision. CONCLUSIONS An aggressive surgical approach did not adversely influence early and late survival following type A aortic dissection; it reduced the necessity of late reoperation and facilitated distal aortic reoperation.

Late reoperation for proximal aortic and arch complications after previous composite graft replacement in Marfan patients.

BACKGROUND Marfan patients who received composite graft replacement for proximal aortic disease frequently require late reoperation. The initial surgical technique for this lesion remains controversial. METHODS Fourteen Marfan patients who received composite graft replacement for annuloaortic ectasia with or without aortic dissection required late reoperation thorough re-median sternotomy. The techniques used for an initial composite graft replacement were the original Bentall procedure in 11 patients, the Cabrol procedure in 2, and coronary button technique in 1. Reoperation was indicated for prosthesis-related complications in 10 patients, distal aortic lesion in 13, or for both lesions in 8. Reoperations were performed, on average, 8.4 years after an initial operation. Reoperative procedures included re-composite graft replacement in 1 patient, total arch replacement in 5, and re-composite graft replacement with total arch replacement in 8. RESULTS There were two in-hospital deaths (14.3%). Although pseudoaneurysms of the coronary artery or distal aorta occurred in the original Bentall or Cabrol procedures, true aneurysms of the coronary artery were noted even in the coronary button technique. Six patients required a total of eight subsequent descending or thoracoabdominal aortic replacements for an aneurysmal formation of a distal false lumen. CONCLUSIONS The coronary button technique, with a small side hole for coronary anastomosis, is the procedure of choice for annuloaortic ectasia because it reduces the risk of coronary artery-related complications. Concomitant total arch replacement may be recommended for annuloaortic ectasia with DeBakey type I aortic dissection in selected patients to avoid the risk of reoperation on the aortic arch.

Cerebral oxygenation monitoring for total arch replacement using selective cerebral perfusion

BACKGROUND This study was undertaken to verify the safety of our total arch replacement assisted by selective cerebral perfusion with respect to cerebral oxygenation. METHODS Subjects to be evaluated were selected between February 1999 and March 2000 and comprised 13 patients who underwent total arch replacement (TAR) (TAR group) and 18 patients who had undergone coronary artery bypass grafting or valve replacement (control group). They were monitored throughout the operation by two-channel near-infrared spectroscopy. Changes in intracranial oxyhemoglobin and the tissue oxygenation index were compared between the two groups. Additionally, jugular venous oxygen saturation was simultaneously measured in 10 patients from each group. Maximum changes in these variables from baseline in the TAR group were compared with those in the control group. Bilateral oxygenation differences between two hemispheres were also evaluated. RESULTS There was no incidence of postoperative cerebral infarction, and no significant difference was observed in the maximum decrease in these variables between the two groups. Bilateral oxygenation differences between the two hemispheres in the TAR group were similar to those in the control group, except for the tissue oxygenation index in the rewarming phase. CONCLUSIONS From the standpoint of cerebral oxygenation, our technique of total arch replacement was nearly as safe as an ordinary cardiac operation.

Interleukin-1 receptor antagonist attenuates the severity of spinal cord ischemic injury in rabbits

OBJECTIVE Thoracic and thoracoabdominal aortic surgery is sometimes complicated by subacute or delayed paraplegia. Pro-inflammatory cytokine interleukin-1 (IL-1) beta has been implicated in extensive inflammation and progressive neurodegeneration after ischemia. Using a rabbit model, we investigated the neuroprotective effects of IL-1 receptor antagonist (IL-1ra) in a temporal fashion. METHODS Spinal cord ischemia was induced by aortic cross-clamping in New Zealand White rabbits. The animals were assigned to three groups. Group C (n = 20) received saline (0.2-mL) and Group I (n = 20) received IL-1ra (200-microg/0.2-mL) intrathecally just after reperfusion. Group S (n = 3) underwent sham operation without aortic occlusion. We assessed the neuroprotective effects of IL-1ra by evaluating neurological function, histopathological changes, and in-situ terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL staining). We also measured the levels of Nitric Oxide (NO) and S100beta in cerebrospinal fluid (CSF). Each evaluation was performed sequentially within 120 hours after reperfusion. RESULTS Group C showed progressive deterioration of motor function which became statistically significant from 48 hours after the onset of reperfusion (P < .05, P < .01, P < .001, P < .001 at 48, 72, 96, and 120 hours, respectively). Compared to Group C, a higher number of viable neurons was observed with less severe spinal cord injury in Group I (P < .01, .05 and .05 at 24, 72, and 120 hours, respectively). TUNEL-positive neurons were also significantly reduced by the administration of IL-1ra (P <.01 and .05 at 24, and 120 hours, respectively). The difference between Group C and Group I with regard to NO was significant at 72 and 120 hours (P < .05), while that in terms of S100beta was significant only at 24 hours (P < .05). CONCLUSIONS Administration of IL-1ra attenuates spinal cord ischemic-reperfusion injury as evidenced by reducing both neuronal necrosis and apoptosis.

Histological changes in canine aorta 1 year after stent-graft implantation: implications for the long-term stability of device anchoring zones.

Purpose: To examine aortic histology 1 year after stent-graft implantation in a canine model as a means of assessing the durability of endograft fixation. Methods: Fourteen mongrel dogs each received 1 stent-graft and 1 bare stent placed endoluminally in the abdominal aorta. Eight animals were followed for 1 year, 3 for 4 to 8 weeks, and the remaining 3 for 24 to 48 hours. Aortic specimens were stained with hematoxylin-eosin, elastica-van Gieson, and Massons trichrome and examined with light and electron microscopy and immunohistochemistry to identify smooth muscle cells (SMC), endothelialization, aortic wall ultrastructure, and changes at the device anchoring sites. Results: No dilatation or dissection was noted at any of the device anchoring sites. The aortic media at 1 year was remarkably decreased in thickness: 891 ± 196 µm in the control tissue versus 388 ± 70 µm for the proximal stent-grafted aorta and 457 ± 148 µm for the bare-stented aortic segment. Other important histological features were reduced elastic lamellae in the stent-grafted aorta versus control (p<0.0001), increased SMC density in the stent-grafted aortic region (p<0.0001 versus control), and absence of inflammatory infiltrate. Complete neointimal covering and endothelialization of the luminal endograft surface were found. SMCs generally showed no ultrastructural features of necrosis. Conclusions: Aortic stent-grafts induce distinctive histological changes in the aortic wall at 1 year, even when implanted in a healthy aorta. Although there is considerable medial elastin loss, an increased medial SMC density, an exuberant neointima, and a general absence of perigraft inflammation suggest an ongoing process of structural restoration at the device anchoring sites.

Changes in cerebrospinal fluid and blood lactate concentrations after stent-graft implantation at critical aortic segment: a preliminary study,

OBJECTIVES Obstruction of blood flow through the arteria radicularis magna (ARM) has been linked with ischemic spinal cord injury after conventional thoracic aortic repair. Whether or not endoluminal stent-grafts, deliberately positioned against this artery can cause similar damage to the spinal cord has not been comprehensively investigated. The purpose of this study was to assess the blood and cerebrospinal fluid (CSF) concentrations of lactate - a well-known biochemical marker of ischemic neurological injury, before and after stent-graft implantation against the ARM. MATERIALS AND METHODS Endoluminal stent-grafting was performed in ten mongrel dogs. In five animals (experimental group), stent-grafts covered the fourth and fifth lumbar segmental arteries - which have been described as the canine equivalents to the ARM in humans. In the remaining five animals (control group), devices of similar length were placed in the lower thoracic aorta. CSF was obtained by cisternal puncture technique at the following time points; before stent-grafting, and 15, 30 and 60 min after stent-grafting. Parallel arterial blood samples were also obtained using a heparinized syringe. All samples were centrifuged and the supernatant analysed for lactate. RESULTS The mean preprocedural lactate concentration in the CSF was 1.7+/-0.3 mmol/l. Mean postprocedural levels in the experimental group at 15, 30 and 60 min were 3.1+/-1.9, 3.9+/-1.1 and 11.9+/-2.5 mmol/l, respectively (control values; 2.1+/-1.9, 2.7+/-1.1 and 1.9+/-1.5 mmol/l, respectively). Mean preprocedural blood lactate level was 1.8+/-0.6 mmol/l, while the mean postprocedural concentrations in the experimental group at 15, 30 and 60 min were 2.9+/-1.2, 3.4+/-1.7 and 3.9+/-2.0 mmol/l, respectively. Two out of the five animals in the experimental group suffered mild to moderate hind limb weakness. CONCLUSION Selective placement of stent-grafts against the ARM in dogs resulted in a conspicuous increase in CSF and blood lactate concentrations 60 min after the procedure with or without physical signs of neurological deficits. Although the small sample size of this preliminary study does not allow any definitive conclusion, it may be worthwhile to confirm the findings in appropriately controlled larger studies.