| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;82:2154-2160
© 2006 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Emory University, Atlanta, Georgia
b Department of Perfusion Services, Emory Affiliated Hospitals, Atlanta, Georgia
Accepted for publication July 7, 2006.
* Address correspondence to Dr Chen, Emory University Division of Cardiac Surgery, 1365 Clifton Rd, Building A, Atlanta, GA 30322 (Email: edward.chen{at}emoryhealthcare.org).
Presented at the Fifty-second Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 10–12, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
METHODS: Over 21 months, SCP through right axillary cannulation with a side graft was performed in 61 patients. Thirty-three percent (20 of 61) underwent emergent operation for Stanford type A dissection or intramural hematoma, including 3 of 20 (4.7%) with pericardial tamponade; the remainder of SCP (41 of 61) was elective. The mean follow-up was 9.1 ± 0.40 months.
RESULTS: Selective antegrade cerebral perfusion was used in 20 of 22 emergent cases (91%), with 2 unsuccessful cannulation attempts, and no peripheral arterial dissections encountered. The SCP flows averaged 16.3 ± 0.71 cc · kg–1 · min–1 for a mean perfusion period of 26.1 ± 1.9 minutes. The average cardiopulmonary bypass time for all patients was 173 ± 11 minutes. Average hospital stay was 8.1 ± 0.80 days. One case (1.3%) of permanent and 3 cases (4.8%) of temporary neurologic dysfunction occurred in SCP patients. The hospital mortality rate for emergent SCP cases (2 of 20, 10%) was not statistically different from the mortality rate for elective SCP cases (3 of 41, 7.3%, p = not significant), with no difference in complication rates. All 3 SCP patients with preoperative tamponade survived without complication. Cerebral oximetry data showed a trend toward decreased left-sided (contralateral) scalp perfusion. There was no association of emergent status with neurologic dysfunction, death, or any other adverse outcome.
CONCLUSIONS: Axillary cannulation and SCP in the surgical treatment of proximal aortic pathology is safe in both elective and emergent settings.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Technique
Anesthetic induction was relatively standard, with sodium pentothal (50 to 100 mg intravenously), isofluorane (1.5% to 3% minimum alveolar concentration), and standard muscle relaxant. Axillary artery exposure was achieved through incision in the right deltopectoral groove, with careful identification and avoidance of injury to the brachial plexus. After administration of 5,000 units heparin sodium, a side-graft of 8-mm Gelweave (Vascutek; Terumo, Ann Arbor, Michigan) was sewn to the axillary artery with running 6-0 polypropylene; this graft was then cannulated with a 24F William Harvey-type arterial cannula (Medtronic, Minneapolis, Minnesota). Venous cannulation was uniformly undertaken with a two-stage cannula (Medtronic) in the right atrial appendage. The innominate artery was snared using umbilical tape and tourniquet, proximal to the right subclavian artery takeoff, to allow for antegrade perfusion of the right common carotid artery.
Patients were cooled to less than 22°C (bladder), with an inflow temperature to the brain of no greater than 18°. When target temperature was reached, the innominate tourniquet was snared down and goal antegrade SCP flow began at flow rates of 10 cc · kg–1 · min–1 and adjusted to maintain mean cerebral perfusion pressures of 60 mm Hg. Myocardial protection was provided with cold blood (66:1) cardioplegia, usually beginning with intermittent retrograde doses, followed by intermittent antegrade doses. All distal aortic anastomoses were performed open during the period of SCP. After the completion of the distal aortic anastomosis, systemic rewarming was initiated with removal of tourniquet and resumption of full-body perfusion. Proximal aortic work as well as any coronary and valve procedures were performed during the rewarming period. In a small subset of patients (11 of 61), transcutaneous cerebral oximetry (INVOS 3100-SD; Troy, Michigan) was monitored through the frontal scalp bilaterally. Values were measured numerous times before bypass, before, during, and after SCP, and after bypass (see Fig 1).
|
2 above this value. Continuous data were initially evaluated with two-sample t tests. If the equality of variances assumption was met, then the pooled t test was employed; otherwise, the Satterthwaite method for unequal variances was used to test for statistical differences. Univariate analysis was then performed with numerous preoperative and postoperative data variables, but a multivariate analysis was not undertaken owing to small sample sizes. Statistical significance was reached at a p value of less than 0.05. |
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
respectively. Regarding preoperative morbidities, preexisting neurologic dysfunction was defined as a history of stroke or transient ischemic attack; chronic renal insufficiency as serum creatinine of 1.5 mg/dL or greater or prior dialysis requirement; chronic obstructive pulmonary disease as either significant (50 pack-years or more) smoking history, actively followed by a pulmonologist for this diagnosis, or significant use of bronchodilators, home oxygen or home continuous positive airway pressure or BiPAP (bilevel positive airway pressure breathing assistance) machine. The average age was 57 ± 2.4 years, and the mean left ventricular ejection fraction was 53% ± 1.6%. The emergent group was significantly more obese and had a much higher male predominance. Procedures (Table 4) are defined as follows: "ascending aortic replacement" included any ascending or hemiarch (in which a beveled portion of the aortic lesser curvature was replaced), whereas "arch replacement" signified the reimplantation of one or more great vessels, either singly or as an island. Approximately half of elective procedures, but the large majority of emergent procedures, involved replacement of ascending aortas. In a large percentage of cases, however, the aortic valves required intervention, including 46% of elective cases (19 of 41) and 55% of emergent cases (11 of 20). That included the root replacements, in which 2 of 8 of elective native valves (25%) and 100% of emergent valves (4 of 4) were spared.
|
|
|
Adverse events are listed in Table 5; respiratory insufficiency was defined as either requiring reintubation, or prolongation of intensive care unit stays of greater than 2 days, from respiratory causes. Renal insufficiency was classified as the acute elevation of serum creatinine greater than 50%, or more than one point, relative to preoperative baseline. There was no significant difference in postoperative temporary neurologic dysfunction in elective and emergent groups: (2 of 41, or 4.9%, elective versus 1 of 20, or 5%, emergent). There was a single stroke, which occurred in the emergent group (1 of 20, 5%). In this latter case, computed tomography identified a distal left middle cerebral artery embolic infarction, as well as several left occipital infarctions representing either left posterior cerebral artery emboli, or watershed territory injury; hence, the etiology was either strictly embolic, or less likely a combination of embolic and contralateral watershed injury. Finally, the hospital mortality rates were not statistically different between groups. In the elective cohort (3 of 41, 7.3%), 1 patient died 8 hours postoperatively of intractable coagulopathy and acidosis after 274 minutes of cardiopulmonary bypass and 219 minutes of cross-clamp time; another died on postoperative day 2 with multisystem failure; and a third died acutely of respiratory arrest on postoperative day 14. Deaths in the emergent group (2 of 20, 10%) were due to anastomotic rupture on postoperative day 9 in 1 patient, and 9 days after surgery in another patient due to mesenteric ischemia and gangrenous bowel. There was found to be no statistical correlation between any of the preoperative comorbidities and postoperative outcomes.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
With the clear sensitivity of the brain to hypoxia, the notion of safe cerebral protection with adequate surgical exposure was found to be feasible with selective perfusion through direct great vessel cannulation. Seminal work in canines by Tanaka and colleagues [14] in 1995 described adequate cerebral protection, as measured by preservation of somatosensory evoked potentials and by minimal histologic changes, when perfusion of all arch vessels was maintained at levels measured to be 50% of physiologic cerebral flow in dogs. This flow rate correlated with 39.8 mm Hg carotid artery pressure, 5.2 cc · kg–1 · min–1 cerebral flow, 60% sagittal sinus blood oxygen saturation, at 25°C (cerebral) for 90 minutes. The authors translated this to 10 cc · kg–1 · min–1 SCP for their clinical human patients, validated by a 1.3% CVA rate in 73 patients, and this flow rate has become a standard for use in proximal aortic surgical procedures [1, 2, 4, 15]. As for the technique of cannulating the axillary artery, several reports in the past 20 years have described the ease of use of bilateral axillary arteries, and their surprisingly low incidence of involvement by dissection [6, 16–18]. Reuthebuch and colleagues [19] demonstrated in 120 patients that the use of unilateral, right subclavian perfusion in acute aortic dissection operations allowed for shorter selective cerebral flow times, at warmer temperatures, with lower morbidity and mortality rates, when compared with femoral perfusion combined with retrograde superior vena caval cerebral protection.
Our study demonstrates, first, that right axillary artery cannulation may be safely applied almost uniformly to a wide variety of proximal aortic pathology, in both elective and emergent settings. Although not the first collection of patients cannulated solely through the right axillary artery for brain perfusion through the right common carotid artery [16], in these patients true DHCA was not necessary, with SCP time a low 26.1 ± 2.3 minutes, at a mild bladder temperature of 23.2°C ± 1.6°C. Our outcomes have been favorable to date, with 3 of 61 (4.9%) transient and 1 of 61 (1.6%) permanent neurologic injuries, and 5 of 61 (8.2%) 30-day mortality. Many of the common complications encountered during this type of perfusion technique—inability to cannulate, or flow sufficiently, through the axillary artery, or brachial plexus injury—are either minimized or absent in this series. We were unable to find significant associations between many traditional risk factors and death, for example, concomitant coronary artery bypass graaft surgery, preoperative comorbidities, and particularly emergent status or presence of acute dissection.
In this report, the permanent stroke rate of 1.6% is lower than that reported in most series. Our pilot near-infrared spectroscopic scalp capillary saturation monitoring data interestingly show, first, that right-sided cutaneous saturations are essentially perfectly preserved throughout the pre-SCP and post-SCP periods, while there is a nonsignificant trend toward a drop in left-sided saturations (see Fig 1). Second, the drop in cutaneous saturations on the left after SCP never falls below 90% of prebypass baseline, thus never even approaching the critical minimal ratio of 70% described by Higami and colleagues [8] and by others—a value below which neurologic injury in humans is reportedly as high as 60%.
Regardless, there are plenty of data to support, first in animal models, the adequacy of full-brain perfusion through unilateral carotid artery flow [18]; and clinically, there is ample evidence that humans, in whom the incidence of incomplete circle of Willis is 8% or probably much less [21, 22], can be very safely perfused through a unilateral cerebral vessel [6, 21–24]. Few authors advocate any preoperative testing (for example, carotid duplex, cerebral angiography) to verify the suitability and safety of SCP in a particular patient, whereas many (including our group) advocate that the visual confirmation of backbleeding from left carotid and subclavian arteries is sufficient evidence of good brain crossflow [21]. Temporary neurologic dysfunction (confusion, slowness to awaken) is thought to be attributable to inadequate brain protection, whereas strokes result from embolic events [5]. We noted two cases of temporary dysfunction, and one of permanent stroke, the latter likely almost entirely due to embolic debris in an emergent patient, suggesting that the type of perfusion did not contribute to this patient’s disability. Unfortunately, we compiled no oximetry data—owing to the random, pilot nature of the sampling—in patients with neurologic injuries. In general, we may speculate that our low incidence of neurologic injury may be attributable, first, to our very short or nonexistent full DHCA periods, and second, to our mean SCP flows (> 16 cc · kg–1 · min–1) exceeding our goal of 10 cc · kg–1 · min–1, with some initial evidence that, even on the contralateral left scalp, there was no significant or sustained drop in cutaneous saturations. In our series, the mean cerebral flow of approximately 1.4 L/min far exceeded the theoretical minimum of 500 to 700 cc/min, as described above [13].
Limitations
The small sample size and the retrospective nature of this series preclude the type of definitive findings—improved survival and outcomes—that may be made with a large, randomized, prospective trial regarding the use of axillary cannulation, versus other standard or alternative cannulation techniques, in elective and emergent proximal aortic procedures. In addition, our series contained fewer cases of full arch replacements than other recent reports, which likely lessened the complexity of cases on the whole. We did not measure the time required to place a sidegraft to, and then cannulate, the right axillary artery, and therefore cannot comment regarding the applicability of this technique in patients requiring cardiopulmonary resuscitation and emergent sternotomy.
Conclusions
Our study has demonstrated that the uniform use of right axillary artery cannulation is appropriate in any proximal or arch aortic case where cross-clamping is undesirable or contraindicated, and in addition, that use in emergent cases is just as safe and efficacious as in the elective setting. Our low rates of temporary and permanent neurologic dysfunction, although culled from a small sample, may be attributable to the nearly nonexistent need for cerebral hypothermic arrest, and to the short selective perfusion times; and the results seem to validate the scientific evidence that demonstrates that unilateral cerebral flow is very safe at moderate temperatures. Further important work on this issue would include a randomized, prospective trial of SCP versus DHCA, or even SCP through unilateral axillary artery versus multiple arch vessels, comparing preoperative and postoperative neurocognitive variables, clinical outcomes, perfusion times, pressures and temperatures, and data quantifying cerebral flow (cerebral oximetry or magnetic resonance). [20]
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
In this study, Drs Budde and Chen present their cannulation strategy for ascending and arch repairs, which admittedly in their experience is in evolution. They make an argument that axillary artery cannulation by way of a side-arm graft with antegrade cerebral perfusion provides both a safe method for cannulation during emergent proximal aortic repairs as well as an acceptable method for cerebral protection. They report excellent results in 61 patients over almost a 2-year period: 8.2% mortality, 1.4% stroke, a 4.7% temporary neurological dysfunction rate, with no differences between the emergent and elective cases. Mean antegrade perfusion rate was 16 cc · kg–1 · min–1, with a mean perfusion pressure of 70 mm Hg, which is a little bit higher than what was previously reported. The mean body temperature bladder was 23°C, with a mean antegrade cerebral perfusion pressure time of 26 minutes. The authors acknowledge limitations in the study, specifically that it is retrospective and noncontrolled. I am still a little uncertain of the role of antegrade cerebral perfusion over retrograde cerebral perfusion or just profound hypothermic circulatory arrest in view of a relatively short circ arrest time. I do commend the authors for using some form of cerebral monitoring, cerebral oximetry, although later in the series. This is something we believe wholeheartedly in Houston, and we use it in every case to guide us.
I do have two general questions. One is, tell us about your cooling strategy. If you don’t use electroencephalography (EEG) and look for isoelectric nature VAG, when do you stop cooling and when do you start your circulatory arrest period? You mentioned in your article that basically in general you have been warming the patients because you are using an antegrade selective cerebral perfusion, and if you are allowing the patients to warm a little bit longer, are there any risks to that?
Again, I would like to congratulate Dr Budde and Dr Chen for their nice work at Emory.
DR BUDDE: Thanks, Dr Estrera. Our cooling strategy, as you say, has been in evolution. At present our strategy is, number one, the nasopharyngeal temperature needs to be 22°C or less, and at that point we infuse blood no greater than 18°C to the brain.
On the question of monitoring: we do not have multichannel EEG at our institution. We are more familiar with the BIS, the Bispectral Index, especially the anesthesiologists, and when the reading goes to zero, we are comfortable that there is no brain electrical activity. We would prefer to have EEG, but we haven’t gotten that capability yet.
In terms of warming, you are asking if our coolest temperatures are warmer than earlier in the study. The answer is yes. We have found recently that we have been able to stay at 26° to 28°C bladder temperature, instead of the 18° to 21°C from earlier in our study. We start to warm soon after the distal anastomosis is done, and have found that rewarming takes a little under 1 hour, which is not bad in some of these obese patients.
DR RICHARD A. JONAS (Washington, DC): I certainly enjoyed your paper. I have had a long-standing interest in brain protection both under circulatory arrest and in conditions of very reduced flow, which is essentially what you are creating here. Both our animal studies and two prospective randomized trials have demonstrated that both the pH strategy and the hematocrit strategy are very important in terms of brain protection. So I am interested to know whether you used the pH-stat strategy, which reduces cerebral metabolic rate and improves oxyhemoglobin dissociation. So with the pH stat strategy, hemoglobin releases oxygen better in a setting of deep hypothermia when oxyhemoglobin has been very leftward shifted by the hypothermia itself. Hematocrit has also turned out to be extremely important. In fact, the NIH shut our study down in children because there was a marked difference in developmental outcome at 1 year of age when we used a hematocrit of 20 versus 30, where there were better outcomes. I do believe this is applicable in adults, too, where emboli are also important as a cause of brain injury with bypass, because if you hemodilute, the brain will compensate by increasing cerebral blood flow. So you will actually deliver more emboli if you use more dilute blood. So can you tell us about your pH strategy and your hematocrit strategy please? Thank you.
DR BUDDE: Thank you, Dr Jonas. It is a privilege to have you ask me questions. We have followed no strict guidelines regarding pH strategy, and our institution as a whole is actually looking at which approach is best in our adults. We keep our hematocrit at between 20 and 22. I am sorry that I cannot be more specific for you, but thank you for your interest.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. E. Halkos, F. Kerendi, R. Myung, P. Kilgo, J. D. Puskas, and E. P. Chen Selective antegrade cerebral perfusion via right axillary artery cannulation reduces morbidity and mortality after proximal aortic surgery J. Thorac. Cardiovasc. Surg., November 1, 2009; 138(5): 1081 - 1089. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kamiya, K. Kallenbach, D. Halmer, M. Ozsoz, K. Ilg, A. Lichtenberg, and M. Karck Comparison of Ascending Aorta Versus Femoral Artery Cannulation for Acute Aortic Dissection Type A Circulation, September 15, 2009; 120(11_suppl_1): S282 - S286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Khaladj, M. Shrestha, S. Peterss, M. Strueber, M. Karck, M. Pichlmaier, A. Haverich, and C. Hagl Ascending aortic cannulation in acute aortic dissection type A: the Hannover experience Eur. J. Cardiothorac. Surg., October 1, 2008; 34(4): 792 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Malvindi, G. Scrascia, and N. Vitale Is unilateral antegrade cerebral perfusion equivalent to bilateral cerebral perfusion for patients undergoing aortic arch surgery? Interactive CardioVascular and Thoracic Surgery, October 1, 2008; 7(5): 891 - 897. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ogino, H. Sasaki, K. Minatoya, H. Matsuda, H. Tanaka, H. Watanuki, M. Ando, and S. Kitamura Evolving arch surgery using integrated antegrade selective cerebral perfusion: Impact of axillary artery perfusion J. Thorac. Cardiovasc. Surg., September 1, 2008; 136(3): 641 - 649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. P. Urbanski, A. Lenos, J. C. Blume, V. Ziegler, B. Griewing, R. Schmitt, A. Diegeler, and M. Dinkel Does anatomical completeness of the circle of Willis correlate with sufficient cross-perfusion during unilateral cerebral perfusion? Eur. J. Cardiothorac. Surg., March 1, 2008; 33(3): 402 - 408. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Jakob, K. Tsagakis, A. Szabo, I. Wiese, M. Thielmann, and U. Herold Rapid and safe direct cannulation of the true lumen of the ascending aorta in acute type A aortic dissection J. Thorac. Cardiovasc. Surg., July 1, 2007; 134(1): 244 - 245. [Full Text] [PDF]
|
|
|
|
|
|
A.K. Koray and S. Dogan Carotid artery cannulation in aortic surgery J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1392 - 1392. [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 |
