| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;84:38-42
© 2007 The Society of Thoracic Surgeons
a Department of Surgery, Cardiothoracic Surgery Division, St. Louis University Medical Center, St. Louis, Missouri
b Mallinckrodt Institute of Radiology, Washington University, St. Louis, Missouri
Accepted for publication March 9, 2007.
* Address correspondence to Dr Lee, Department of Surgery, Cardiothoracic Surgery Division, 3635 Vista Ave at Grand Blvd, PO Box 15250, St. Louis, MO 63110-0250 (Email: ricklee{at}nmh.org).
Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: The rate of permanent strokes was examined at a single institution between two time periods using the 
2 method. From January 2002 through June 2003 (pre-CT) 230 patients underwent elective cardiac procedures. All patients received routine intraoperative transesophageal echocardiography but only selective epiaortic scanning. From July 2003 through December 2005 (post-CT), 273 patients underwent cardiac operations and 114 patients at high risk for a stroke (history of CVA or transient ischemic attack, peripheral vascular disease, end-stage renal disease, age >70 years old or calcifications on cardiac catheterization) also received a preoperative noncontrast chest CT.
Results: In the pre-CT period, 7 patients received epiaortic scans that resulted in a change in operative strategy in only 1 patient. In the post-CT period, the 114 CT scans identified 20 patients with significant calcifications in the ascending aorta. Operative strategy was altered in 19 patients, including 12 axillary artery cannulations, four off-pump, no-touch aorta coronary bypasses, and three ascending aortic replacements. The CVA rate was 3.04% in the pre-CT period and 0.73% in the post-CT period (p = 0.05).
Conclusions: Aggressive screening and avoidance of ascending aortic atheroma reduces stroke. Preoperative CT scans in patients at high risk may help identify aortic areas at risk before entering the operating room, lead to more thorough screening in the operating room, and result in a more aggressive strategy to avoid calcified areas.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Noncontrast computed tomography (CT) identifies areas of aortic calcification [9]. Identification of aortic calcification on CT is straightforward and should already be in the armamentarium of every thoracic surgeon. This test can also be performed before an elective sternotomy, and potential alterations of operative approach and any revised risk can then be discussed with patients in the preoperative setting.
The aim of this study was to determine if noncontrast CT scans performed on a high-risk population before cardiac surgical procedures could identify areas of aortic calcification that might be a source of intraoperative emboli. Further, we sought to determine the postsurgical incidence of stroke after avoiding manipulation of the identified areas.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
All patients received intraoperative transesophageal echocardiography (TEE). Epiaortic scanning was performed on all patients who had aortic atheroma on TEE or had clinical suspicion for aortic atheroma, such as palpable aortic irregularities or calcium on cardiac catheterization as determined by the surgeon.
Beginning July 2003, we began an aggressive program to preoperatively identify aortic calcification. During this (post-CT) era, 273 patients underwent elective cardiac operations, and 114 patients at high risk for a stroke (history of cerebrovascular accident or transient ischemic attack, peripheral vascular disease, age >70 years old, chronic renal failure or calcifications on cardiac catheterization) received a preoperative noncontrast chest CT in addition to the established echocardiographic protocol. An assessment was made on adequacy of area for cannulation, cross-clamping, and proximal anastomoses (Fig 1 and Fig 2). Calcified areas were avoided.
|
|
Baseline characteristics were compared between the two groups of patients using the 2 test (Table 1). Age and ejection fraction were compared with the Student t test. The rates of CVA and mortality were examined at a single institution between two time periods using the 2 test (Table 2). A multivariate logistic regression was also performed on previously documented predictors of stroke [1].
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Seven epiaortic scans were performed in the 230 patients undergoing cardiac operations in the pre-CT era. In 6 of these 7 patients, the aorta was cannulated in a place free of calcium, as determined by the scan. In 1 patient, an off-pump coronary artery bypass grafting (CABG) was performed without aortic manipulation in which the innominate artery was used a proximal anastomosis.
Seven patients (3%) sustained strokes during the pre-CT era. No postoperative strokes occurred in patients who received an epiaortic scan.
Of the 273 patients undergoing surgery in the post-CT period, 114 patients were thought to be a high risk for stroke and underwent preoperative noncontrast chest CT scan. This identified 20 patients with significant calcifications in the ascending aorta. Operative strategy was altered in 19 patients, including 12 axillary artery cannulations, four off-pump no-touch CABGs, and three ascending aortic replacements. At times, this also allowed for aortic cannulation in calcium-free areas identified by the scan, but the number of times this occurred was not recorded.
Only the 20 patients in whom significant calcium was identified on preoperative CT received epiaortic scans at the time of surgery. No epiaortic scan resulted in an additional change in treatment plan.
Two strokes occurred in the post-CT era. One patient had a history of stroke but did not receive a preoperative CT scan. One patient did receive a preoperative CT scan that identified aortic calcification. The patient had an area of aorta adequate for cross-clamping but insufficient area for both cannulation and cross-clamping. Attempts at cutdowns of the right and left axillary arteries and right and left femoral arteries were met with calcified vessels that could not be cannulated. The patient was deemed too elderly and frail to tolerate an ascending aortic replacement so the operation was conducted with all manipulations on the ascending aorta.
Overall, there was a trend toward a higher stroke rate in the pre-CT period of 3.04% compared with 0.73% the post-CT period (p = 0.05). Mortality was also higher in the pre-CT era at 13.5% compared with 7% in the post-CT era (p = 0.01). All reported permanent strokes were devastating and resulted in acute death. In the population that underwent cardiac operations and did not have a postoperative stroke, the mortality difference loses significance (11.3% versus 6.2%; p = 0.06).
A multivariate logistical regression analysis demonstrated the only significant risk factors for stroke were performing valve surgery (p = 0.007) and the absence of a preoperative CT scan (p = 0.037).
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The value of this study is that it presents a preoperative tool to assess the ascending aorta in high-risk patients. The surgeon can easily identify patients and offer alternate strategies before a sternotomy. In some instances not reported here, it may lead to medical management or percutaneous intervention.
There are many limits to this study. It is a very small retrospective study with a very small number of events. The number of strokes was low in both groups, and only permanent strokes were reported.
Other large patient series have already reported the complete elimination of stroke by using meticulous echo-directed aortic cannulation and perioperative hemodynamic management [6]. The reduction of stroke after cardiac operations by avoiding aortic atheroma is already cardiac surgical dogma [4–8, 11–13].
This study did not examine neurologic changes that were not considered strokes. Previous works have determined that although manipulation of aortic atheroma is a major cause of stroke after cardiac surgery, atheroma does not appear to be the primary factor in the pathogenesis of cognitive changes after cardiac operations [14].
It is somewhat unusual that all the strokes in this report were acute. Although stroke has been reported to occur in an early and late phase after cardiac operations, with up to 65% in the late phase [10], more recent reports suggest that with aggressive detection, most strokes are found shortly after surgery [3, 15]. It is also notable that there was no difference in postoperative atrial fibrillation, which is a potential cause for late stroke when combined with low cardiac output [10].
It is important also to remember that the same individuals collected the data over the same time period and used the same criteria for reporting events during the 3 years. Any potential inconsistency in recording events compared with other studies should be the same inconsistency during both the pre-CT and post-CT eras. This study did not use some of the highly sophisticated tools available for statistical analysis, such as propensity matching. It relied on a limited data set and limited analysis.
It is interesting that this technique appears to preoperatively identify calcification in patients who are at risk for severe atheroma embolization. In addition, the precision that the CT provides allows for limited aortic manipulation. The options for avoiding the calcified plaque can be considerably less extensive, such as using alternate cannulation sites. It also notable that the only two strokes in the post-CT era occurred after either a failure to follow the protocol or failure to use the information provided by the CT.
Further, the mortality is fairly high, especially in the pre-CT era. We believe this reflects the high-risk inner city patient population and have internally confirmed the institutional patient complexity by the National Quality Improvement Project database. Alternately, some of the fatalities may represent strokes or atheroemboli that were not recognized. It is curious, though, that the difference in mortality is eliminated after the patients who sustained strokes are withdrawn. This suggests that avoiding strokes may contribute to improved survival; however, this speculation is clearly beyond the limits of this analysis.
Finally, this study did not assess the risk to patients or the economic impact of adding an additional preoperative test in high-risk patients. A noncontrast CT exposes the patient to a small amount of radiation, but this is relatively benign in the adult population [16]. Although strokes have been reported to increase the cost of cardiac operations fourfold [8], the economic impact of stroke reduction versus the increased cost of preoperative CT was not performed. Invariably, there will be a group of patients in whom the increased cost is not justified. In our own institution, the results of this study have led to an almost uniform application of the preoperative noncontrast CT to adults undergoing cardiac operations and an abandoning of the epiaortic scan in this group. This is most likely an inappropriate use of resources that will need to be streamlined after future analysis that includes other institutions and more patients.
A potential deficit of the noncontrast CT may be a failure to identify noncalcified atheroma as effectively as epiaortic scanning. Experience using noncontrast CT as a screening tool is very limited, so it is unclear if it will be as effective. Only further experience with CT and its comparison with epiaortic scanning will define the comparative value of these two modalities, which may be complementary.
Preoperative imaging studies neither cause nor prevent morbidity or mortality after cardiac operations. They merely provide information to guide the surgeon. Conversely, how the surgeon obtains and uses preoperative information does affect the outcome after the procedure. The more aggressively we look for aortic atheroma and avoid its manipulation, the more likely we are to reduce atheroemboli.
This study provides an application of an existing diagnostic study that helps our group to identify and avoid areas that are known to embolize and cause stroke after cardiac procedures. We hope that other surgeons find this to be a useful strategy in the appropriate patients, as well as a guide to offer a more accurate risk assessment. As newer CT-based noninvasive coronary imaging becomes available, this information may be provided on a routine basis and should be used.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Would there be any additional benefit of using intravenous contrast, especially in the case of anticipated complex reoperative procedures? Secondly, can you assure us that over the course of this study the operative procedures have been conducted similarly with regard to perfusion variables, application of aortic clamps, and the use of venting? And finally, has your experience allowed you to define the specific patient characteristics that warrant obtaining a preoperative CT scan? Thank you.
DR LEE: Thank you for those thoughtful comments and questions. We actually believe there is a disadvantage to intravenous contrast, for two reasons. First, we avoid any possible nephrotoxicity by using no-contrast CTs, and the only major disadvantage to using noncontrast CTs is a little radiation, which is not harmful in the adult population, and the cost associated with it. The second reason is that we believe the contrast impairs our ability to identify calcium. It is possible, but when you get the contrast, there is less differentiation between a calcified aortic wall and intravenous contrast.
As to the second point, we did have multiple surgeons. We had no difference between any of the surgeons and stroke, and all patients did receive systemic mild hypothermic perfusion, single aortic cross-clamping, and aortic root venting with intermittent left ventricular venting, which was also variable depending on individual surgeons and within the individual surgeon’s practice.
And lastly, I think ideally I could answer yes, we have identified criteria that will define the appropriate patients, but I think that is impossible, for a number of reasons. Number one, this is a very small sample, and in order to really answer that question, we need a large series of patients at multiple institutions. In addition, it is really a philosophical question. I can share with you briefly our personal experience. We have got complete buy-in at our institution, so now we order pre-op CT scans, or our cardiology fellows order pre-op CT scans on all our patients undergoing cardiac surgery, regardless of their age. This is probably an inappropriate use of resources; however, we have had a decline in our stroke rate from this series since that time. The question is, how many strokes should you avoid? I believe we should avoid all of them.
DR STEPHEN CASSIVI (Rochester, MN): Thanks, Rick, for your presentation, and I would like to congratulate you and your group, especially your wife Michelle for entering or continuing in the field of improving patient safety. Again, I would like to note that you are probably one of only a few members, including Dr Urschel and Dr Ungerleider, of this great Association who have been able to include their wives in their presentations and their work here.
I would wonder if we can learn a little bit more from what you presented by your describing those protocol violations that led to the two deaths. I think we can maybe learn a little bit what happened with breaking of your usual protocol.
DR LEE: I think that is a great question. In the post-CT era, we had 2 patients who had strokes. One patient had a history of stroke, but we did not get a CAT scan; we missed it. In the other patient—as I mentioned, we were able to avoid the calcium in 19 patients—this twentieth patient was an 80-year-old female who had critical aortic stenosis, who had insufficient room for both aortic cannulation and cross-clamping. Subsequently, she underwent an axillary artery cutdown on the right that we could not cannulate, axillary cutdown on the left, femoral cutdown on the right, femoral cutdown on the left, and we couldn’t get anywhere to cannulate. I decided to cannulate the aorta, and rather than do an aortic replacement, I just proceeded with the operation, and it ended up in a fatal stroke.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
