- Email: [email protected]
A double patch sandwich technique for surgical repair of acute aortic dissection type A
Tatiana M. Fleck, MD, Heinz Tschernich, MD, Martin Grabenwoger, MD, Doris Hutschala, MD, Herbert Koinig, MD, Ernst Wolner, MD, and Werner Mohl, MD, PhD Departments of Cardiothoracic Surgery and Cardiothoracic and Vascular Anaesthesiology, University of Vienna, Vienna, Austria
Background. This study compares two groups of patients with acute aortic dissection type A in whom two different techniques of anastomotic reinforcement were used, and evaluates the impact of these two techniques on perioperative blood loss and surgical outcome. Methods. One hundred eighty-five consecutive patients with acute aortic dissection type A between 1998 and 2002 were grouped according to the technique utilized for reinforcing the aortic anastomotic site. Group A consisted of 21 patients in whom a novel double patch sandwich technique was used, whereas in group B (164 patients) conventional Teflon felt strips served as reinforcement. Results. Preoperative profiles were comparable for both groups. Deep hypothermia and circulatory arrest were used in 180 patients (97%). Retrograde cerebral perfusion and retrograde cardioplegia were used in 70
patients (38%). Hospital mortality for the two groups was 4.7% (1 of 21) and 18% (29 out of 164), p < 0.30, respectively. On average 2 versus 6 U of PRBC (p < 0.21), and 3 versus 5 U of FFP (p < 0.004) were given during operation in groups A and B, respectively, and indicates reduced suture line bleeding in group A. Conclusions. Modifications in the surgical technique for repair of acute aortic dissection type A, and in anastomotic reinforcement in particular, may lead to substantial reduction of suture line bleeding and diminished blood loss and transfusion requirements, and favorably affects patient outcome. Nevertheless, continued effort is mandatory to further enhance surgical outcome in this patient population.
S
Material and Methods
ince 1955, when DeBakey and colleagues [1] reported the first successful resection and graft replacement of the ascending thoracic aorta using cardiopulmonary bypass, operative and perioperative management steadily evolved over the years [1–3]. However, despite these advancements, inhospital mortality remains unacceptably high at 19% to 25%, and stresses the importance for continued improvement in prevention, diagnosis, and treatment of acute aortic dissection [4 –7]. Herein we want to compare our experience with a double patch sandwich technique that reinforces the aortic wall layers and diminishes blood loss in the warming period against a control collective where this technique was not utilized.
Accepted for publication Feb 15, 2003. Address reprint requests to Dr Mohl, Department of Cardiothoracic Surgery, AKH Vienna, Leitstelle 20A, Wa¨hringer Gu¨rtel 18-20, 1090 Vienna, Austria; e-mail: [email protected].
© 2003 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2003;76:499 –502) © 2003 by The Society of Thoracic Surgeons
Patients From September 1997 through August 2002, 185 patients underwent surgery for acute aortic dissection type A at our institution. Patients were grouped according to the technique utilized for aortic anastomosis. All clinical data were obtained by retrospective review of hospital records and personal communication with the patients for follow-up. Group A consisted of 21 patients (female to male ratio 3:18, median age 57 years old, range 37 to 79 years old) operated on by one surgeon (W.M), who introduced a modification of the standard operative technique, using the double patch sandwich technique for reinforcing the aortic wall layers at the anastomotic site, whereas Group B (164 patients, female to male ratio 62:102, median age 57 years old, ⫾ 20 years) served as the control and were operated by six different staff surgeons using conventional Teflon felt strips (Bard PTFE, IMPRA, Tempe, AZ) for reinforcement of the anastomosis. The 164 patients of group B were randomized and were comparable to group A in relation to comorbidities and preoperative profiles. The diagnosis of dissection was established by computed tomography, magnetic resonance imaging, and transoesophageal echocardiography. Hypertension was 0003-4975/03/$30.00 PII S0003-4975(03)00459-4
CARDIOVASCULAR
A Double Patch Sandwich Technique for Surgical Repair of Acute Aortic Dissection Type A
500
FLECK ET AL ANASTOMOSIS AUGMENTATION IN ACUTE AORTIC DISSECTION TYPE A
Ann Thorac Surg 2003;76:499 –502
CARDIOVASCULAR
the most common comorbidity, with a prevalence of 90% in both groups, followed by cardiac disorders, namely coronary artery disease in 26 patients (48%) and congestive heart disease in 31 patients (17%). Ten patients (19%) exhibited Marfan syndrome. A history of previous cardiac surgery was found in 13 patients (7%). The location of the primary entry tear was in the ascending aorta in 166 patients (90%) of the entire cohort and in the aortic arch in 19 patients (12%). Important preoperative complications of aortic dissection sustained by the patients were as follows: free blood in the pericardium was found in 75 patients (41%), cardiac tamponade with hemodynamic instability in 5 patients (12%), and moderate to severe aortic valve insufficiency in 37 patients (20%).
Surgical Technique All operations were performed through a median sternotomy. After patients received heparin (300 U/kg), cardiopulmonary bypass (CPB) was established through the femoral artery and the right atrium. In patients with cardiac tamponade, the femoral vein was used for venous access. Once CPB was instituted, systemic cooling was initiated. After onset of ventricular fibrillation, the aorta was clamped and cold cardioplegia (4°C) was delivered first antegrade and subsequently retrograde in an intermittent manner every 20 minutes in conjunction with topical cooling (ice slush). In group A only retrograde cardioplegia was used. For cerebral protection the patients heads were packed in ice and 1000 mg of methylprednisolone was administered before circulatory arrest. When a rectal temperature of 18°C and nasopharyngeal temperature of 11°C was reached, hypothermic circulatory arrest was initiated. Retrograde cerebral perfusion (RCP) was implemented through the superior vena cava and a mean central venous pressure of 20 mm Hg (flow rate median 280 mL/min) was maintained during the period of circulatory arrest. During the period of circulatory arrest retrograde cardioplegia was continuously given as described above. Albumin precoated polyethylene Dacron grafts (Vaskutek; Sulzer Medica Inc, Renfrewshire, Scotland), with a premanufactured 8-mm sidearm, were used in all patients.
Anastomotic Technique Group A During cooling and after aortic clamping, the dissected aorta was partly removed, the aortic valve was resuspended, and the site of the anastomosis was prepared as follows. A 1-cm wide strip of bovine glutaraldehyde fixed pericardium (W.L. Gore and Associates, Evry Cedex, France) was wrapped over the dissected layers of the aortic wall, forming a concealed cap. This cap was reinforced on the outside with a 0.5-cm wide Teflon felt (Bard PTFE, IMPRA; Fig 1A and 1B). A continuous 4-0 Prolene suture (Ethicon, Sommerville, NJ) was used in a mattress-suture technique to reinforce the aortic wall layers. Thereafter the prosthesis was sutured to the sandwich with a 4-0 Prolene running suture. Care was taken to suture the entire sandwich, which was facilitated by the
Fig 1. Schematic drawings of the sandwich construction. (A) The dissected aortic wall layers are reinforced on the inside and outside with the pericard cap. (B) An additional reinforcement is made with a Teflon felt strip on the outside.
use of a bigger needle (Ethicon SH). No additional glue technique was used to fill the dissected space. During circulatory arrest the same technique was used to cover the space between the dissected layers on the distal anastomosis. After this reinforcement of the anastomotic site, the graft was sutured in a standard fashion using continuous 4-0 Prolene. After removing all air from the circulation extracorporeal circulation was reestablished through the side branch of the aortic graft. After reaching a rectal temperature of 36.5°C, CPB was discontinued and the canulas were removed.
Anastomotic Technique Group B The distal anastomosis was performed in a conventional manner with continuous 4-0 Prolene and reinforced on the outside with a Teflon felt strip. After the distal anastomosis was finished, the arterial side graft of the CPB was connected to the 8-mm sidearm of the prosthesis and selective antegrade perfusion and rewarming of the patient was initiated. Meanwhile, the proximal anastomosis was sutured with continuous 4-0 Prolene after reinforcing the aortic wall with Teflon felt strips.
Statistical Analysis Data were analyzed by Anova univariate analysis. Results are displayed as median values with standard
FLECK ET AL ANASTOMOSIS AUGMENTATION IN ACUTE AORTIC DISSECTION TYPE A
Table 1. Univariate Analysis of Blood and Coagulation Variables Median Value Variable PRBC FFP RBC POD 1 Hemoglobin POD 1 Hematokrit POD 1 Platelets POD 1 apTT POD 1 TZ POD 1 Revision bleeding
A
B
p Value
3.76 units 3.38 units 3.67 T/L 11.00 g/dL 31.8% 108 G/L 40.7 seconds 84% 0.09 (2/21)
4.71 units 5.57 units 3.2 T/L 9.55 g/dL 28% 96 G/L 50.0 seconds 79% 0.23 (31/164)
⬍ 0.21 ⬍ 0.0043 ⬍ 0.007 ⬍ 0.0017 ⬍ 0.0051 ⬍ 0.31 ⬍ 0.28 ⬍ 0.51 ⬍ 0.22
Platelets, 1 unit ⫽ 200 mL. apTT ⫽ activated thromboplastin time; FFP ⫽ fresh frozen plasma, 1 unit ⫽ 200 mL; Hb ⫽ hemoglobin; Hk ⫽ hematokrit; NT ⫽ Normotest; POD ⫽ postoperative day; PRBC ⫽ packed red blood cells, 1 unit ⫽ 250 mL; Rev ⫽ revision.
deviation. A p value of less than 0.05 was considered significant.
Results Operative Data Significant differences between the two groups were detected in respect to median operating times (350 minutes [⫾ 117 minutes] versus 437 minutes [⫾ 373 minutes], p ⬍ 0.001) and aortic cross-clamp times (median 99 minutes [⫾ 55 minutes] versus 122 minutes [⫾ 75 minutes], p ⬍ 0.04) for groups A and B, respectively. Extracorporeal circulation times (median 195 minutes [⫾ 150 minutes] and 202 minutes [222 minutes], p ⬍ 0.75) and DHCA times (median 26 minutes [⫾ 15 minutes] versus 31 minutes [⫾ 8 minutes], p ⬍ 0.31) did not reach statistical significance. Concomitant coronary artery bypass grafting was performed in 1 patient (4%) in group A and in 10 patients (6%) in group B. A composite graft with reimplantation of the coronary arteries was utilized in 4 patients of group A (19%) and 26 patients of group B (16%). Hemiarch repair was performed in 5 patients (24%) of group A and 87 patients (53%) in group B. Concomitant procedures were coronary artery bypass grafting in 11 patients (6%), and open stent graft placement into the descending aorta in 8 patients (4%). Intraoperatively, on average, 3 U (⫾ 3,6) of red blood cells (RBC) and 4 U (⫾ 4.2) of FFP were administered in group A; whereas, in group B, 5 U (⫾ 8.4) of RBC and 6 U of FFP (⫾ 6,7) were administered (p ⬍ 0.21, p ⬍ 0.004). Significant differences between the two groups were apparent in the amount of reduction of blood cells and consumption of hemostatic factors on postoperative day 1 by comparable values preoperatively. For more details, refer to Table 1. On postoperative day 1 the following laboratory values were found: univariate analysis revealed significant differences in total blood count (median 3.2 T/L vs 3.6 T/L,
501
p ⬍ 0.007), hemoglobin (median 11.0 g/dL vs 9.5 g/dL, p ⬍ 0.001), and hematocrit (median 31.8% vs 28%, p ⬍ 0.005). Slight differences, but not statistically significant, could be noted in the coagulation system as elucidated in Table 1. These findings correlated with the amount of chest tube drainage in the first 24 hours after operation, with a median of 800 mL in group A in contrast to 1400 mL in group B, and indicates a 45% reduction of blood loss in group A.
Hospital Outcome Overall hospital mortality for groups A and B was 5% (1/21) and 18% (29/164), with a p value of less than 0.30. Causes were as follows: uncontrollable bleeding intraoperatively in 3 patients, multiorgan failure in 12 patients, septic shock in 3 patients, stroke in 3 patients, and heart failure in 9 patients. The incidence of reexploration for bleeding was significantly lower in group A (10%, 2/21) versus group B (19%, 31/164), p less than 0.22. Postoperative permanent cerebral complications caused by embolic stroke occurred in 12 patients (1 patient from group A, 11 patients from group B). Temporary neurologic dysfunction, defined as the occurrence of postoperative confusion, and transient delirium was present in 2 patients (10%) of group A and in 26 patients (16%) of group B.
Follow-Up Average long-term follow-up for the entire cohort was approximately 36 months, ranging from 1 to 59 months. Routine follow-up consisted of a 6-week postoperative visit at our outpatient department to ensure optimal blood pressure control. Then the patients were examined every 6 months with computed tomographic scans in the first year after operation and yearly thereafter except patients with Marfan syndrome, who continued with 6-month follow-up intervals.
Comment The characteristics of acute aortic dissection are the separation of the layers within the aortic wall with blood entering the intima-media space and, therefore, enhancing dissection. Usually one or more entry tears in the intima layer allow communication between the two lumen, thereby promoting mal perfusion of the lower extremity and end organ ischemia when perfused by the false lumen. The current mortality rates, recently outlined by Bachet and colleagues [4], have been estimated to be between 9% and 15% for patients with acute aortic dissection type A. These improved results, in contrast to former reports [1, 5], are attributed mainly to advances in early diagnosis as well as improvements in surgical techniques and perioperative management. Despite advances in surgical experience and technology, operations for ascending aortic dissections are still associated with substantial risk, particularly when carried out in critically ill patients.
CARDIOVASCULAR
Ann Thorac Surg 2003;76:499 –502
502
FLECK ET AL ANASTOMOSIS AUGMENTATION IN ACUTE AORTIC DISSECTION TYPE A
CARDIOVASCULAR
Whenever an intima tear was located in the aortic arch we considered it mandatory to replace the arch and reimplant the arch vessels as an island into the prosthesis. The aortic valve was preserved whenever possible, except in patients with Marfan syndrome where we preferred to replace the aortic root. These recommendations are strongly supported by several reports with evidence of fewer reoperations when the arch is respected at the presence of an intima tear and a conservative approach regarding the native aortic valve to avoid prolongation of operating time and the associated increase in hospital mortality [6, 8, 9]. So far only a limited number of studies have focused on the different techniques used for anastomotic reinforcement, such as a study reported by Sabik and coworkers in 2000 [10], where the proximal false lumen was obliterated with a tailored felt strip and two further felt stripes were placed inside and outside of the aorta. The dissected aortic wall layers and the three layers of felt strip were then sandwiched together with monofilament sutures. It is suggested that obliteration of the false lumen with the Teflon felt strip is sufficient reinforcement, therefore Bio glue (CryoLife Inc., Kennesaw, GA) that is used to reapproximate the dissected aortic wall layers, is rarely used at our department and we are aware of the associated risk of tissue necrosis [11]. Excessive bleeding after aortic surgery is generally related to a combination of several alterations in the hemostatic system pertaining to the dilution and activation of the coagulation system, which is mainly attributed to the use of extracorporeal circulation. Our study evaluated a technique producing a cap covering the dissected layers reinforced on the outside by a Teflon felt. With this method the blood loss through the intermediate dissected space can be prevented. Furthermore, we noticed reduced suture bleeding due to the sealing effect of the pericardium cap. Hemostasis was achieved in a shorter time period, thereby reducing operating time and the amount of blood derivates. The additional use of antifibrinolytics was implemented in November 2002, so these adjuncts were not used in the study population. Laboratory values on postoperative day 1 revealed a marked difference between the two groups regarding hemoglobin values and coagulation factor levels, which leads to the assumption that the pericardium cap effectively prevents excessive bleeding from the anastomotic site. These findings were further supported by the fact that chest tube drainage differed significantly, between 500 mL and 900 mL for groups A and B, respectively. Most of these observed beneficial effects on patient outcome can
Ann Thorac Surg 2003;76:499 –502
be explained as consequences of a reduced intraoperative and perioperative blood loss [12]. The longer intensive care unit and hospital stays of group A can be explained by the introduction of fast track protocols in the last 2 years, reducing the overall median stay, whereas most of the patients of group A were operated on from 1997 to 1999. Therefore, we can conclude that our method of anastomotic reinforcement may have the ability to reduce intraoperative blood loss and to decrease operating times, and thereby favorably affect patient outcome after surgery for acute aortic dissection type A. Limitations of the study were the retrospective collection of data and the associated considerations as the learning curves from surgeons, as well as improvements in intraoperative and postoperative management over a 5-year period. This has to be taken into account and may have influenced clinical outcome.
References 1. DeBakey M, Cooley D, Creech O Jr. Surgical considerations of dissecting aneurysm of the aorta. Ann Surg 1955;142:586 – 612. 2. Kouchoukos NT, Dougenis D. Surgery of the thoracic aorta. N Engl J Med 1997;336:1876 –88. 3. Fann JI, Smith JA, Miller DC, et al. Surgical management of aortic dissection during a 30 year period. Circulation 1995; 92(9 Suppl):II1 13–21. 4. Bachet J, Goudot B, Dreyfus GD, et al. Surgery for acute type A aortic dissection: The hospital Foch experience. Ann Thorac Surg 1999;67:2006 –9. 5. Ehrlich MP, Ergin A, McCullough JN, et al. Results of immediate surgical treatment of all acute type A dissections. Circulation 2000;102[Suppl III]:III-248 –52. 6. Shapira OM, Aldea GS, Cutter SM, Fitzgerald CA, Lazar HL, Shemin RJ. Improved clinical outcomes after operation of the proximal aorta: a 10-year experience. Ann Thorac Surg 1999;67:1030 –7. 7. Hagan PG, Nienhaber CA, Isselbacher EM, et al. The international registry of acute aortic dissection: new insights into an old disease. JAMA 2000;283:897–903. 8. Westaby S, Saito S, Katsumata T. Acute type A dissection: conservative methods provide consistently low mortality. Ann Thorac Surg 2002;73:707–13. 9. Bachet J. Acute type A aortic dissection: Can we dramatically reduce the surgical mortality? Ann Thorac Surg 2002;73: 701–3. 10. Sabik JF, Lytle BW, Blackstone EH, McCarthy PM, Loop FD, Cosgrove DM. Long term effectiveness of operations for ascending aortic dissections. J Thorac Cardiovasc Surg 2000; 119:946 –62. 11. Kazui T, Washiyama N, Hasan A, et al. Role of biologic glue repair of the proximal aortic dissection in the development of early and midterm redissection of the aortic root. Ann Thorac Surg 2001;72:509 –14. 12. Despotis GJ, Avidan MS, Hogue CW. Mechanisms and attenuation of haemostatic activation during extracorporal circulation. Ann Thorac Surg 2001;72:S1821–31.
Background. This study compares two groups of patients with acute aortic dissection type A in whom two different techniques of anastomotic reinforcement were used, and evaluates the impact of these two techniques on perioperative blood loss and surgical outcome. Methods. One hundred eighty-five consecutive patients with acute aortic dissection type A between 1998 and 2002 were grouped according to the technique utilized for reinforcing the aortic anastomotic site. Group A consisted of 21 patients in whom a novel double patch sandwich technique was used, whereas in group B (164 patients) conventional Teflon felt strips served as reinforcement. Results. Preoperative profiles were comparable for both groups. Deep hypothermia and circulatory arrest were used in 180 patients (97%). Retrograde cerebral perfusion and retrograde cardioplegia were used in 70
patients (38%). Hospital mortality for the two groups was 4.7% (1 of 21) and 18% (29 out of 164), p < 0.30, respectively. On average 2 versus 6 U of PRBC (p < 0.21), and 3 versus 5 U of FFP (p < 0.004) were given during operation in groups A and B, respectively, and indicates reduced suture line bleeding in group A. Conclusions. Modifications in the surgical technique for repair of acute aortic dissection type A, and in anastomotic reinforcement in particular, may lead to substantial reduction of suture line bleeding and diminished blood loss and transfusion requirements, and favorably affects patient outcome. Nevertheless, continued effort is mandatory to further enhance surgical outcome in this patient population.
S
Material and Methods
ince 1955, when DeBakey and colleagues [1] reported the first successful resection and graft replacement of the ascending thoracic aorta using cardiopulmonary bypass, operative and perioperative management steadily evolved over the years [1–3]. However, despite these advancements, inhospital mortality remains unacceptably high at 19% to 25%, and stresses the importance for continued improvement in prevention, diagnosis, and treatment of acute aortic dissection [4 –7]. Herein we want to compare our experience with a double patch sandwich technique that reinforces the aortic wall layers and diminishes blood loss in the warming period against a control collective where this technique was not utilized.
Accepted for publication Feb 15, 2003. Address reprint requests to Dr Mohl, Department of Cardiothoracic Surgery, AKH Vienna, Leitstelle 20A, Wa¨hringer Gu¨rtel 18-20, 1090 Vienna, Austria; e-mail: [email protected].
© 2003 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2003;76:499 –502) © 2003 by The Society of Thoracic Surgeons
Patients From September 1997 through August 2002, 185 patients underwent surgery for acute aortic dissection type A at our institution. Patients were grouped according to the technique utilized for aortic anastomosis. All clinical data were obtained by retrospective review of hospital records and personal communication with the patients for follow-up. Group A consisted of 21 patients (female to male ratio 3:18, median age 57 years old, range 37 to 79 years old) operated on by one surgeon (W.M), who introduced a modification of the standard operative technique, using the double patch sandwich technique for reinforcing the aortic wall layers at the anastomotic site, whereas Group B (164 patients, female to male ratio 62:102, median age 57 years old, ⫾ 20 years) served as the control and were operated by six different staff surgeons using conventional Teflon felt strips (Bard PTFE, IMPRA, Tempe, AZ) for reinforcement of the anastomosis. The 164 patients of group B were randomized and were comparable to group A in relation to comorbidities and preoperative profiles. The diagnosis of dissection was established by computed tomography, magnetic resonance imaging, and transoesophageal echocardiography. Hypertension was 0003-4975/03/$30.00 PII S0003-4975(03)00459-4
CARDIOVASCULAR
A Double Patch Sandwich Technique for Surgical Repair of Acute Aortic Dissection Type A
500
FLECK ET AL ANASTOMOSIS AUGMENTATION IN ACUTE AORTIC DISSECTION TYPE A
Ann Thorac Surg 2003;76:499 –502
CARDIOVASCULAR
the most common comorbidity, with a prevalence of 90% in both groups, followed by cardiac disorders, namely coronary artery disease in 26 patients (48%) and congestive heart disease in 31 patients (17%). Ten patients (19%) exhibited Marfan syndrome. A history of previous cardiac surgery was found in 13 patients (7%). The location of the primary entry tear was in the ascending aorta in 166 patients (90%) of the entire cohort and in the aortic arch in 19 patients (12%). Important preoperative complications of aortic dissection sustained by the patients were as follows: free blood in the pericardium was found in 75 patients (41%), cardiac tamponade with hemodynamic instability in 5 patients (12%), and moderate to severe aortic valve insufficiency in 37 patients (20%).
Surgical Technique All operations were performed through a median sternotomy. After patients received heparin (300 U/kg), cardiopulmonary bypass (CPB) was established through the femoral artery and the right atrium. In patients with cardiac tamponade, the femoral vein was used for venous access. Once CPB was instituted, systemic cooling was initiated. After onset of ventricular fibrillation, the aorta was clamped and cold cardioplegia (4°C) was delivered first antegrade and subsequently retrograde in an intermittent manner every 20 minutes in conjunction with topical cooling (ice slush). In group A only retrograde cardioplegia was used. For cerebral protection the patients heads were packed in ice and 1000 mg of methylprednisolone was administered before circulatory arrest. When a rectal temperature of 18°C and nasopharyngeal temperature of 11°C was reached, hypothermic circulatory arrest was initiated. Retrograde cerebral perfusion (RCP) was implemented through the superior vena cava and a mean central venous pressure of 20 mm Hg (flow rate median 280 mL/min) was maintained during the period of circulatory arrest. During the period of circulatory arrest retrograde cardioplegia was continuously given as described above. Albumin precoated polyethylene Dacron grafts (Vaskutek; Sulzer Medica Inc, Renfrewshire, Scotland), with a premanufactured 8-mm sidearm, were used in all patients.
Anastomotic Technique Group A During cooling and after aortic clamping, the dissected aorta was partly removed, the aortic valve was resuspended, and the site of the anastomosis was prepared as follows. A 1-cm wide strip of bovine glutaraldehyde fixed pericardium (W.L. Gore and Associates, Evry Cedex, France) was wrapped over the dissected layers of the aortic wall, forming a concealed cap. This cap was reinforced on the outside with a 0.5-cm wide Teflon felt (Bard PTFE, IMPRA; Fig 1A and 1B). A continuous 4-0 Prolene suture (Ethicon, Sommerville, NJ) was used in a mattress-suture technique to reinforce the aortic wall layers. Thereafter the prosthesis was sutured to the sandwich with a 4-0 Prolene running suture. Care was taken to suture the entire sandwich, which was facilitated by the
Fig 1. Schematic drawings of the sandwich construction. (A) The dissected aortic wall layers are reinforced on the inside and outside with the pericard cap. (B) An additional reinforcement is made with a Teflon felt strip on the outside.
use of a bigger needle (Ethicon SH). No additional glue technique was used to fill the dissected space. During circulatory arrest the same technique was used to cover the space between the dissected layers on the distal anastomosis. After this reinforcement of the anastomotic site, the graft was sutured in a standard fashion using continuous 4-0 Prolene. After removing all air from the circulation extracorporeal circulation was reestablished through the side branch of the aortic graft. After reaching a rectal temperature of 36.5°C, CPB was discontinued and the canulas were removed.
Anastomotic Technique Group B The distal anastomosis was performed in a conventional manner with continuous 4-0 Prolene and reinforced on the outside with a Teflon felt strip. After the distal anastomosis was finished, the arterial side graft of the CPB was connected to the 8-mm sidearm of the prosthesis and selective antegrade perfusion and rewarming of the patient was initiated. Meanwhile, the proximal anastomosis was sutured with continuous 4-0 Prolene after reinforcing the aortic wall with Teflon felt strips.
Statistical Analysis Data were analyzed by Anova univariate analysis. Results are displayed as median values with standard
FLECK ET AL ANASTOMOSIS AUGMENTATION IN ACUTE AORTIC DISSECTION TYPE A
Table 1. Univariate Analysis of Blood and Coagulation Variables Median Value Variable PRBC FFP RBC POD 1 Hemoglobin POD 1 Hematokrit POD 1 Platelets POD 1 apTT POD 1 TZ POD 1 Revision bleeding
A
B
p Value
3.76 units 3.38 units 3.67 T/L 11.00 g/dL 31.8% 108 G/L 40.7 seconds 84% 0.09 (2/21)
4.71 units 5.57 units 3.2 T/L 9.55 g/dL 28% 96 G/L 50.0 seconds 79% 0.23 (31/164)
⬍ 0.21 ⬍ 0.0043 ⬍ 0.007 ⬍ 0.0017 ⬍ 0.0051 ⬍ 0.31 ⬍ 0.28 ⬍ 0.51 ⬍ 0.22
Platelets, 1 unit ⫽ 200 mL. apTT ⫽ activated thromboplastin time; FFP ⫽ fresh frozen plasma, 1 unit ⫽ 200 mL; Hb ⫽ hemoglobin; Hk ⫽ hematokrit; NT ⫽ Normotest; POD ⫽ postoperative day; PRBC ⫽ packed red blood cells, 1 unit ⫽ 250 mL; Rev ⫽ revision.
deviation. A p value of less than 0.05 was considered significant.
Results Operative Data Significant differences between the two groups were detected in respect to median operating times (350 minutes [⫾ 117 minutes] versus 437 minutes [⫾ 373 minutes], p ⬍ 0.001) and aortic cross-clamp times (median 99 minutes [⫾ 55 minutes] versus 122 minutes [⫾ 75 minutes], p ⬍ 0.04) for groups A and B, respectively. Extracorporeal circulation times (median 195 minutes [⫾ 150 minutes] and 202 minutes [222 minutes], p ⬍ 0.75) and DHCA times (median 26 minutes [⫾ 15 minutes] versus 31 minutes [⫾ 8 minutes], p ⬍ 0.31) did not reach statistical significance. Concomitant coronary artery bypass grafting was performed in 1 patient (4%) in group A and in 10 patients (6%) in group B. A composite graft with reimplantation of the coronary arteries was utilized in 4 patients of group A (19%) and 26 patients of group B (16%). Hemiarch repair was performed in 5 patients (24%) of group A and 87 patients (53%) in group B. Concomitant procedures were coronary artery bypass grafting in 11 patients (6%), and open stent graft placement into the descending aorta in 8 patients (4%). Intraoperatively, on average, 3 U (⫾ 3,6) of red blood cells (RBC) and 4 U (⫾ 4.2) of FFP were administered in group A; whereas, in group B, 5 U (⫾ 8.4) of RBC and 6 U of FFP (⫾ 6,7) were administered (p ⬍ 0.21, p ⬍ 0.004). Significant differences between the two groups were apparent in the amount of reduction of blood cells and consumption of hemostatic factors on postoperative day 1 by comparable values preoperatively. For more details, refer to Table 1. On postoperative day 1 the following laboratory values were found: univariate analysis revealed significant differences in total blood count (median 3.2 T/L vs 3.6 T/L,
501
p ⬍ 0.007), hemoglobin (median 11.0 g/dL vs 9.5 g/dL, p ⬍ 0.001), and hematocrit (median 31.8% vs 28%, p ⬍ 0.005). Slight differences, but not statistically significant, could be noted in the coagulation system as elucidated in Table 1. These findings correlated with the amount of chest tube drainage in the first 24 hours after operation, with a median of 800 mL in group A in contrast to 1400 mL in group B, and indicates a 45% reduction of blood loss in group A.
Hospital Outcome Overall hospital mortality for groups A and B was 5% (1/21) and 18% (29/164), with a p value of less than 0.30. Causes were as follows: uncontrollable bleeding intraoperatively in 3 patients, multiorgan failure in 12 patients, septic shock in 3 patients, stroke in 3 patients, and heart failure in 9 patients. The incidence of reexploration for bleeding was significantly lower in group A (10%, 2/21) versus group B (19%, 31/164), p less than 0.22. Postoperative permanent cerebral complications caused by embolic stroke occurred in 12 patients (1 patient from group A, 11 patients from group B). Temporary neurologic dysfunction, defined as the occurrence of postoperative confusion, and transient delirium was present in 2 patients (10%) of group A and in 26 patients (16%) of group B.
Follow-Up Average long-term follow-up for the entire cohort was approximately 36 months, ranging from 1 to 59 months. Routine follow-up consisted of a 6-week postoperative visit at our outpatient department to ensure optimal blood pressure control. Then the patients were examined every 6 months with computed tomographic scans in the first year after operation and yearly thereafter except patients with Marfan syndrome, who continued with 6-month follow-up intervals.
Comment The characteristics of acute aortic dissection are the separation of the layers within the aortic wall with blood entering the intima-media space and, therefore, enhancing dissection. Usually one or more entry tears in the intima layer allow communication between the two lumen, thereby promoting mal perfusion of the lower extremity and end organ ischemia when perfused by the false lumen. The current mortality rates, recently outlined by Bachet and colleagues [4], have been estimated to be between 9% and 15% for patients with acute aortic dissection type A. These improved results, in contrast to former reports [1, 5], are attributed mainly to advances in early diagnosis as well as improvements in surgical techniques and perioperative management. Despite advances in surgical experience and technology, operations for ascending aortic dissections are still associated with substantial risk, particularly when carried out in critically ill patients.
CARDIOVASCULAR
Ann Thorac Surg 2003;76:499 –502
502
FLECK ET AL ANASTOMOSIS AUGMENTATION IN ACUTE AORTIC DISSECTION TYPE A
CARDIOVASCULAR
Whenever an intima tear was located in the aortic arch we considered it mandatory to replace the arch and reimplant the arch vessels as an island into the prosthesis. The aortic valve was preserved whenever possible, except in patients with Marfan syndrome where we preferred to replace the aortic root. These recommendations are strongly supported by several reports with evidence of fewer reoperations when the arch is respected at the presence of an intima tear and a conservative approach regarding the native aortic valve to avoid prolongation of operating time and the associated increase in hospital mortality [6, 8, 9]. So far only a limited number of studies have focused on the different techniques used for anastomotic reinforcement, such as a study reported by Sabik and coworkers in 2000 [10], where the proximal false lumen was obliterated with a tailored felt strip and two further felt stripes were placed inside and outside of the aorta. The dissected aortic wall layers and the three layers of felt strip were then sandwiched together with monofilament sutures. It is suggested that obliteration of the false lumen with the Teflon felt strip is sufficient reinforcement, therefore Bio glue (CryoLife Inc., Kennesaw, GA) that is used to reapproximate the dissected aortic wall layers, is rarely used at our department and we are aware of the associated risk of tissue necrosis [11]. Excessive bleeding after aortic surgery is generally related to a combination of several alterations in the hemostatic system pertaining to the dilution and activation of the coagulation system, which is mainly attributed to the use of extracorporeal circulation. Our study evaluated a technique producing a cap covering the dissected layers reinforced on the outside by a Teflon felt. With this method the blood loss through the intermediate dissected space can be prevented. Furthermore, we noticed reduced suture bleeding due to the sealing effect of the pericardium cap. Hemostasis was achieved in a shorter time period, thereby reducing operating time and the amount of blood derivates. The additional use of antifibrinolytics was implemented in November 2002, so these adjuncts were not used in the study population. Laboratory values on postoperative day 1 revealed a marked difference between the two groups regarding hemoglobin values and coagulation factor levels, which leads to the assumption that the pericardium cap effectively prevents excessive bleeding from the anastomotic site. These findings were further supported by the fact that chest tube drainage differed significantly, between 500 mL and 900 mL for groups A and B, respectively. Most of these observed beneficial effects on patient outcome can
Ann Thorac Surg 2003;76:499 –502
be explained as consequences of a reduced intraoperative and perioperative blood loss [12]. The longer intensive care unit and hospital stays of group A can be explained by the introduction of fast track protocols in the last 2 years, reducing the overall median stay, whereas most of the patients of group A were operated on from 1997 to 1999. Therefore, we can conclude that our method of anastomotic reinforcement may have the ability to reduce intraoperative blood loss and to decrease operating times, and thereby favorably affect patient outcome after surgery for acute aortic dissection type A. Limitations of the study were the retrospective collection of data and the associated considerations as the learning curves from surgeons, as well as improvements in intraoperative and postoperative management over a 5-year period. This has to be taken into account and may have influenced clinical outcome.
References 1. DeBakey M, Cooley D, Creech O Jr. Surgical considerations of dissecting aneurysm of the aorta. Ann Surg 1955;142:586 – 612. 2. Kouchoukos NT, Dougenis D. Surgery of the thoracic aorta. N Engl J Med 1997;336:1876 –88. 3. Fann JI, Smith JA, Miller DC, et al. Surgical management of aortic dissection during a 30 year period. Circulation 1995; 92(9 Suppl):II1 13–21. 4. Bachet J, Goudot B, Dreyfus GD, et al. Surgery for acute type A aortic dissection: The hospital Foch experience. Ann Thorac Surg 1999;67:2006 –9. 5. Ehrlich MP, Ergin A, McCullough JN, et al. Results of immediate surgical treatment of all acute type A dissections. Circulation 2000;102[Suppl III]:III-248 –52. 6. Shapira OM, Aldea GS, Cutter SM, Fitzgerald CA, Lazar HL, Shemin RJ. Improved clinical outcomes after operation of the proximal aorta: a 10-year experience. Ann Thorac Surg 1999;67:1030 –7. 7. Hagan PG, Nienhaber CA, Isselbacher EM, et al. The international registry of acute aortic dissection: new insights into an old disease. JAMA 2000;283:897–903. 8. Westaby S, Saito S, Katsumata T. Acute type A dissection: conservative methods provide consistently low mortality. Ann Thorac Surg 2002;73:707–13. 9. Bachet J. Acute type A aortic dissection: Can we dramatically reduce the surgical mortality? Ann Thorac Surg 2002;73: 701–3. 10. Sabik JF, Lytle BW, Blackstone EH, McCarthy PM, Loop FD, Cosgrove DM. Long term effectiveness of operations for ascending aortic dissections. J Thorac Cardiovasc Surg 2000; 119:946 –62. 11. Kazui T, Washiyama N, Hasan A, et al. Role of biologic glue repair of the proximal aortic dissection in the development of early and midterm redissection of the aortic root. Ann Thorac Surg 2001;72:509 –14. 12. Despotis GJ, Avidan MS, Hogue CW. Mechanisms and attenuation of haemostatic activation during extracorporal circulation. Ann Thorac Surg 2001;72:S1821–31.