| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ann Thorac Surg 1997;63:9-11
© 1997 The Society of Thoracic Surgeons
Department of Cardiac and Thoracic Surgery, Lahey Hitchcock Clinic, Burlington, Massachusetts
The history of aortic surgery, particularly for brain and spinal cord protection, is punctuated by the oscillations of the pendulum of vogue surgical procedures. Initial hypothermic circulatory arrest or distal perfusion techniques methods advocated separately by Bigelow, Pontius, Barnard, and Borst [1–5] in the 1950s and early 1960s were later largely abandoned because of the associated high mortality rates, risks, and bleeding complications. For arch repairs, selective brain perfusion techniques were then espoused as improvements in cardiopulmonary bypass techniques and technology enabled better control, and because cooling and rewarming was less prolonged, the time on pump was shorter with fewer systemic complications. Later, however, there was a general disillusionment with selective brain perfusion techniques via the brachiocephalic vessels because of the high risk of neurologic complications. Crawford and Saleh [6] reported that 25% of patients died, and four of the five deaths (80%) were due to stroke. This resulted in
the search for a simpler technique and reexamination of the use of deep hypothermic circulatory arrest. Thus, in the late 1970s, Griepp and associates [7], Cooley's group [8], and Crawford's group [6] became strong advocates of deep hypothermic circulatory arrest. Even then, the method was noted to be associated with an increased bleeding risk and rate of blood transfusion. By January 1991, Crawford's group [9] had accumulated a series of 656 patients who had undergone deep hypothermic circulatory arrest with a 10% mortality rate and 7% risk of a neurologic deficit. In the absence of a prospective, randomized study, it is this comprehensive series against which other brain protection techniques are compared.
In this issue, Hayashi and colleagues [10] report a study from nine centers of 143 patients who underwent aortic arch repairs for nondissecting aortic disease. As expected from a retrospective, multicenter analysis, the data are incomplete and operative repairs varied, but nonetheless some important issues are raised. Of note, 15 patients had ruptured aneurysms, which strongly influenced the mortality and stroke rate (27% stroke rate). All patients with ruptured aneurysms, as shown by Kaplan-Meier analysis, died within 3 months of the operation. Of interest is the reexamination of the technique of selective antegrade perfusion at moderate hypothermia in 128 patients. Of these 128 patients, 18 (14.1%) died within 30 days and 8.6% had neurologic deficit. This multicenter study compares favorably with the report from Crawford's group [9] in which the respective incidences of death and neurologic injury for the 380 patients with nondissecting aneurysms were 7.6% and 6.1%. Has the pendulum swung again in favor of selective perfusion? The answer is a qualified no. Clearly, the use of deep hypothermia and circulatory arrest is preferable in the following situations: acute dissection (fragile aorta and dissection up into the brachiocephalic arteries), loose atheroma in the aortic arch (manipulations or threading of cannulas through the arteries will result in embolization), when no perfusion cannulas are wanted in the field (the operative repair of the aortic arch is simpler without them as noted by our predecessors), and when continuous retrograde brain perfusion is used (simultaneous antegrade and retrograde may cause brain edema). Furthermore, in comparison, patients operated on using retrograde brain perfusion with deep hypothermic circulatory arrest have been reported to have neurologic injury rates of 0% to 2%. As first described by Mills and Oschner [11] in 1980 and noted in the initial experience with 50 cases from Crawford's group [9], one of the most important benefits of retrograde brain perfusion is the flushing out of the arch of air or embolic material. Has the pendulum settled on a combined approach? Antegrade selective perfusion could be used during initial total body circulatory arrest followed by retrograde brain perfusion at the termination of the period of total body circulatory arrest to flush out embolic material. The major factor militating against this would be the complexity of the operation and longer circulatory arrest periods, as has been noted with continuous retrograde brain perfusion. With our current level of knowledge, deep hypothermia cannot also be dispensed with when prolonged periods of possible brain ischemia are anticipated. In summary, as shown previously in the study of 656 patients [9], the total cardiopulmonary bypass time and the patient characteristics, such as rupture and renal dysfunction (the latter also noted in the study by Hayashi and colleagues [10]), are better independent predictors of brain injury. Which, then, is the best method of brain protection during aortic arch repairs? Ideally, a prospective, randomized study evaluating postoperative neurocognitive function should be done to determine which method is best. In the meantime, it would appear that deep hypothermic arrest with retrograde brain perfusion is the most easily reproducible and best method in most patients provided the central venous perfusion pressure does not exceed 30 to 35 mm Hg and the circulatory arrest time does not extend beyond 40 to 50 minutes.
As with the vacillations of opinions on the protection of the brain, the pendulum is again swinging in favor of distal aortic perfusion techniques with hypothermia for the protection of organs during descending thoracic aortic or thoracoabdominal aortic operations. What has been the impetus for this? The reasons are the unacceptably high incidences of paraplegia and renal failure in many centers after these operations following aortic cross-clamping alone, the improved technology in bypass circuits (many of the operations can now be done without homologous blood transfusions [12]; thus bleeding complications, initially reported from leading centers as resulting in mortality rates of 25% to 33% from uncontrolled bleeding with bypass circuits, are no longer as great a problem, even with deep hypothermia and circulatory arrest), and the use of distal perfusion methods that allow for safe organ perfusion, temperature control, and fluid administration [13]. Indeed, numerous reports from diverse centers are again supporting the use of distal aortic perfusion based on the lower risk of renal failure and spinal cord injury, although none have been reported as prospective, randomized studies [13–19].
In this issue, Hayashi and colleagues [20] report on 120 patients operated on in nine centers using distal aortic perfusion with either femorofemoral partial cardiopulmonary bypass, atriofemoral bypass, or shunts, in an attempt to determine which distal perfusion method is best. They conclude that all three methods are acceptable, but this requires closer scrutiny. In their series, all neurologic events (three brain events, one paraplegia) occurred after the use of shunts or partial cardiopulmonary bypass. The overall incidence is probably an underestimate because this is a retrospective, multicenter review. Of note, Verdant, who used to champion the use of shunts and reported no paraplegia or significant renal failure in 366 patients [18], has changed to the use of atriofemoral bypass with a centrifugal pump (A. Verdant, personal communication, June 1995) because of a mortality rate of 12% and a high stroke rate. Which, then, is the preferable technique, atriofemoral bypass or femorofemoral partial cardiopulmonary bypass? The answer is both, depending on the operative indications. In 832 patients undergoing descending thoracic aortic repairs, the mortality, renal failure, and neurologic injury rates for atriofemoral bypass in 247 patients were 4%, 4%, and 6% respectively [19]. For cardiopulmonary bypass in 28 patients, the respective rates were 11%, 11%, and 7%, and for cross-clamping alone, 9%, 8%, and 5%. These mortality and renal failure rates were significantly different (p < 0.05). Furthermore, if the cross-clamp time exceeded 40 minutes, then by logistic regression analysis, atriofemoral bypass was also significantly protective. Clearly, there is a role for both partial cardiopulmonary bypass and atriofemoral bypass, with personal preferences also playing a role, particularly the increasing use of femorofemoral partial cardiopulmonary bypass by vascular surgeons. I currently prefer atriofemoral bypass with a heat exchanger and systemically cooling the patient to 30°C before doing a sequential and segmental repair of the aorta in most patients [13]. If, however, the descending aortic repair is more complex, for example, because of reoperation, scar tissue, contained rupture, distal aortic arch involvement, bronchoesophageal fistulas, or pulmonary artery involvement, then femorofemoral arteriovenous partial cardiopulmonary bypass is sometimes preferable because of the greater ease of conversion to deep hypothermic circulatory arrest as long as the aorta has not been cross-clamped [13].
We must always be prepared to evaluate our preconceptions and what we have been taught to keep pace with evolving technology and operative techniques to "best discern and investigate the truth" in taking care of our patients. Curiously, central nervous system protection has reverted to techniques advocated 40 years ago, although technology and operative repair of the aorta have improved considerably over the ensuing period. Much still needs to be done to improve the morbidity of these extensive operations, particularly when the entire aorta, thoracic aorta, or thoracoabdominal aorta has to be replaced, especially as we come under greater scrutiny for the cost of these expensive and demanding operations.
Footnotes
Address reprint requests to Dr Svensson, Department of Cardiac and Thoracic Surgery, Lahey Hitchcock Clinic, 41 Mall Rd, Burlington, MA 01805.
"It profits one who is pondering on the movement, pulsation, performance, function, and services of the heart and arteries to read what his predecessors have written ...and to note the general trend of opinion and confirm their correct statements and through ...manifold experiments and persistent careful observation amend their own ones ...to discern and investigate the truth."William Harvey, introduction, De Motu Cordis, 1628
References
Related Article
This article has been cited by other articles:
|
|
 |
|
  K. Minatoya, H. Ogino, H. Matsuda, H. Sasaki, H. Tanaka, J. Kobayashi, T. Yagihara, and S. Kitamura Evolving Selective Cerebral Perfusion for Aortic Arch Replacement: High Flow Rate With Moderate Hypothermic Circulatory Arrest Ann. Thorac. Surg., December 1, 2008; 86(6): 1827 - 1831. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Ueda, Y. Okita, S. Aomi, H. Koyanagi, and S. Takamoto Retrograde cerebral perfusion for aortic arch surgery: analysis of risk factors Ann. Thorac. Surg., June 1, 1999; 67(6): 1879 - 1882. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Dossche, M. A.A.M. Schepens, W. J. Morshuis, F. E. Muysoms, J. J. Langemeijer, and F. E.E. Vermeulen Antegrade selective cerebral perfusion in operations on the proximal thoracic aorta Ann. Thorac. Surg., June 1, 1999; 67(6): 1904 - 1910. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. R. Juca Operation on the thoracic aorta Ann. Thorac. Surg., August 1, 1998; 66(2): 606 - 607. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
