Ann Thorac Surg 2005;79:1945-1949
© 2005 The Society of Thoracic Surgeons
Original article: Cardiovascular
Newly Developed Aortic Dissection in the Abdominal Aorta After Femoral Arterial Perfusion
Kazumasa Orihashi, MD*,
Taijiro Sueda, MD,
Kenji Okada, MD,
Katsuhiko Imai, MD
Division of Cardiovascular Surgery, Hiroshima University Hospital, Minami-ku, Hiroshima, Japan
Accepted for publication January 3, 2005.
* Address reprint requests to Dr Orihashi, Division of Cardiovascular Surgery, Hiroshima University Hospital, Kasumi 1-2-3, Minami-ku, Hiroshima 734-8551, Japan (E-mail: orichan{at}hiroshima-u.ac.jp).
 |
Abstract
|
|---|
BACKGROUND: Aortic dissection after femoral arterial perfusion has been reported as a dreadful complication. Apart from such a drastic event, using intraoperative transesophageal echocardiography we detected aortic dissection confined to the abdominal aorta that was associated with visceral malperfusion.
METHODS: We examined 11 consecutive patients with aortic dissection in whom the abdominal aorta was intact, and surgeries were performed with femoral perfusion. The abdominal aorta and visceral branches were examined for new development of dissection or malperfusion by means of transesophageal echocardiography before, during, and after cardiopulmonary bypass. These echocardiographic findings were then related to the postoperative assessment and midterm results.
RESULTS: Aortic dissection was found in 3 of 11 cases (27.2%). Unusual progression of metabolic acidosis (base excess 
10 mEq/L) occurred, possibly as a result of malperfusion of visceral arteries, in 2 of these 3 cases, whereas none of the 8 cases presented with such findings. The presence of dissection was later confirmed by postoperative computed tomography in all but 1 case. In the midterm follow-up period, aneurysm formation was found in the infrarenal aorta and iliac arteries in 2 of the 3 cases with new aortic dissection but not in any of the remaining 8 cases.
CONCLUSIONS: New development of aortic dissection after femoral arterial perfusion was found in 27% of the cases in this series. Although these occurred without dramatic symptoms, this event may be related to unusual metabolic acidosis during cardiopulmonary bypass or to subsequent aneurysmal formation of the infrarenal aorta or iliac arteries, or both.
|
Introduction
|
|---|
Aortic dissection after femoral arterial perfusion is a rare but dreadful complication [1, 2]. It can be fatal when it leads to rupture of the aorta or malperfusion of the major branch arteries. Unfortunately, it is impossible to predict the occurrence of dissection. Axillary arterial perfusion can also be used as the primary route of perfusion [3, 4], but is not free from complications [3]. There are also occasions where the axillary artery is not a suitable route for arterial perfusion, and the femoral artery is used as the second choice.
Aortic dissection becomes apparent when it extends proximally to the portion of aorta that is visible within the surgical field. Apart from such an advanced form of dissection, there may be a proportion of aortic dissection that is less extensive and remains undiagnosed during surgery. There have been no reports on the latter type of dissection because no appropriate diagnostic measures have been feasible in the operating theater.
To minimize the perioperative complications related to the various events that occur in the blind zone during cardiovascular surgery, we have used intraoperative transesophageal echocardiography (TEE) and established a technique of visualizing the abdominal aorta and visceral branch arteries [5, 6]. Using this method, we have experienced several cases of newly developed aortic dissection after femoral arterial perfusion that were confined to the abdominal aorta. In this paper, we report the incidence, TEE findings, and subsequent prognosis during a midterm follow-up period.
|
Patients and Methods
|
|---|
We examined 11 consecutive cases of patients who underwent surgery for aortic dissection with cardiopulmonary bypass using femoral arterial perfusion between 1996 and 2000, including 5 cases surgically treated in the acute stage (within 14 days from onset) and 6 cases treated in the chronic stage (later than the 15th day from onset). Fourteen cases during the same period were excluded from the analysis because the abdominal aorta or visceral branch arteries, or both, were already involved at the time of surgery. The cases after 2001 were also excluded from the study because the arterial perfusion route was changed to the axillary artery. These cases included 6 men and 5 women, with ages ranging from 35 to 84 years. These aortic dissections were DeBakey types I, II, and III in 4, 5, and 2 cases, respectively. Surgery was performed in the chronic stage in 2 cases of type III dissection and in 1 case of type II dissection. The surgical procedures included replacement of ascending aorta (including aortic root replacement), ascending aorta to arch, and descending aorta in 8, 1, and 2 cases, respectively. Surgeries were performed by a single surgeon (the second author) during this period.
The femoral artery was surgically exposed and carefully cannulated with an arterial cannula (USCI cannula, 18F or 20F, model 1857; Bard Inc, Billerica, MA) through an arteriotomy (in the right or left femoral artery in 5 and 6 cases, respectively). A venous cannula was placed in the femoral vein, or a two-stage cannula in the right atrium. The heart-lung machine was operated by a single perfusionist during this period. The systemic arterial pressure was monitored continuously in the right or left radial artery. Blood gas analysis was performed every 10 minutes during cardiopulmonary bypass.
A 5MHz biplane TEE (EUB-555; Hitachi, Tokyo, Japan) was initiated after induction of anesthesia and was used for routine intraoperative monitoring and additionally to examine the abdominal aorta. This study was approved by the Institutional Ethics Committee on Human Research, and informed consent was obtained from every patient. The TEE was conducted by a single echocardiographer (the first author) throughout this period. The abdominal aorta and visceral arteries were visualized before, at the onset, during, and after cardiopulmonary bypass.
The abdominal aorta was visualized as reported previously [5, 6]. From the level for scanning of the thoracic descending aorta, the TEE probe was advanced into the stomach with a counterclockwise rotation of the probe to keep the aorta in view. The celiac artery appeared first at the 12 to 2 o’clock position relative to the aorta. As the probe was advanced further, the celiac artery disappeared, and the superior mesenteric artery appeared as the second branch in the same direction as the aorta. As the probe was advanced further, the renal arteries were occasionally able to be visualized. The abdominal aorta and visceral branch arteries were examined for dissection (presence of an intimal flap) in two-dimensional views and for malperfusion (absence of flow signal) in color Doppler mode.
The results of surgeries were assessed by means of computed tomography and angiography or magnetic resonance imaging, as necessary. Postoperative assessment was provided by the presiding radiologist who was blinded to the intraoperative TEE findings. The patients were followed up in the clinic of our department.
|
Results
|
|---|
The celiac and superior mesenteric arteries and the abdominal aorta could be clearly visualized to the level of these arteries in every case during surgery, and no complications related to manipulations of the TEE probe in the stomach were encountered. By inspection, there was no dissection at the femoral arterial cannulation site in any case. New development of aortic dissection was detected in the abdominal aorta in 3 of 11 cases (27.2%): 1 of 5 cases with acute dissection and 2 of 6 cases with chronic dissection. There were no apparent unusual data in the conventional monitoring nor any events of unusual elevation of perfusion pressure during femoral perfusion in these cases except for metabolic acidosis (described in detail below). There was no operative mortality or in-hospital death during this series, and the clinical course during and immediately after surgery was uneventful in every case. The follow-up period ranged from 49 to 98 months (with an average of 67.8 months).
Case 2 was a 62-year-old man with chronic DeBakey type III dissection, who underwent replacement of the descending thoracic aorta under mild hypothermia with aortic cross-clamping. Preoperative computed tomography showed that the visceral branch arteries and infrarenal portion of aorta were intact. Until the initiaion of femoral arterial perfusion, blood flow in the celiac, superior mesenteric, and renal arteries were clearly detected with TEE (Fig 1A). One hour later, however, the flow in the celiac artery became undetectable (Fig 1B), whereas the blood flow in the superior mesenteric artery and renal arteries remained detectable. An intimal flap became apparent in the abdominal aorta. The base excess was -6.4 mEq/L at this time. When antegrade perfusion was resumed, the intimal flap in the aorta was thrust toward the posterior wall and the flow in the celiac artery became detectable, but with a new dissection apparent (Fig 1C). The dissection extended distally to the peripheral branches of the celiac artery (Fig 1D). Base excess gradually became normalized by the end of surgery, and the postoperative course was uneventful. Although postoperative computed tomography did not examine the aorta at the level of the visceral arteries, postoperative angiography demonstrated no unusual findings in the abdominal aorta and iliac arteries. Three years later, however, dissection became apparent below the level of the renal arteries and the internal iliac arteries, with aneurysmal changes in the common iliac arteries. Computed tomography 7 years after the surgery revealed dilatation of the infrarenal abdominal aorta (Fig 2B), bilateral iliac arteries (Fig 2C), and common iliac arteries (Fig 2D). There was no aortic dissection at the level of the renal artery (Fig 2A). The patient subsequently underwent replacement of the abdominal aorta and iliac arteries.
View larger version (98K):
[in this window]
[in a new window]
|
Fig 1. Transesophageal echocardiogram demonstrating the malperfusion and new development of dissection in celiac artery (CEA) in case 2. (A) Before bypass. (B) During bypass; the CEA flow became undetectable. (C) After bypass; the CEA flow recovered but with dissection in the CEA and aorta (AO). Intimal flap is seen. (D) In the peripheral portion of CEA, dissection extends distally into the branch arteries.
|
|
View larger version (135K):
[in this window]
[in a new window]
|
Fig 2. Computed tomogram after 7 years of surgery in case 2. (A) No dissection at the level of right renal artery. (B) Infrarenal aorta with dilated false lumen. (C) False lumen in the internal iliac artery has expanded. (D) Dissection in the left common iliac artery and aneurysm of right common iliac artery.
|
|
Case 1 was an 82-year-old man with chronic DeBakey type III dissection, who underwent replacement of the descending thoracic aorta. Preoperative computed tomography showed no dissection in the abdominal aorta. Before bypass, the blood flow in the visceral arteries was clearly detectable with TEE. Femoral arterial perfusion was initiated uneventfully, and the body temperature was reduced to 20°C. During this period, however, the blood flow signal in the celiac artery became undetectable, whereas that in the aorta remained detectable. An intimal flap was then observed in the abdominal aorta. The surgeon was immediately informed of these findings. Absence of blood flow in the celiac artery was confirmed by epiaortic echography. The base excess was around -7 mEq/L. Under circulatory arrest and retrograde cerebral perfusion, open proximal anastomosis, reconstruction of three intercostal arteries, and open distal anastomosis were performed. False lumen in the thoracic aorta ended above the level of the celiac artery. The distal portions of the aorta were not visible from the surgical field. When antegrade perfusion was resumed, blood flow in the celiac artery became detectable but with new dissection with a narrowed true lumen. The base excess was as low as -17.2 mEq/L, but it gradually became normalized during the rewarming period. The immediate postoperative course was uneventful, without symptoms or signs of intestinal ischemia. Plain computed tomography at 4 months after surgery demonstrated unusual posterior bulging of abdominal aorta at the level of visceral arteries. (It should be noted that the patient refused the use of contrast media at that time.) Contrast computed tomography at 6 years after surgery revealed aortic dissection in the abdominal aorta at the level of the superior mesenteric artery (Fig 3A) and new aneurysmal changes in the infrarenal aorta (50 mm in diameter) and iliac arteries (Fig 3B). The patient declined further surgical treatment.
View larger version (139K):
[in this window]
[in a new window]
|
Fig 3. Computed tomogram after 6 years of surgery in case 1. (A) New dissection is present at the level of superior mesenteric artery (SMA). (B) Infrarenal portion of aorta shows aneurysmal change (50 mm in diameter).
|
|
Case 11 was a 68-year-old man with acute type I dissection who was scheduled for replacement of the ascending aorta. Preoperative computed tomography and intraoperative TEE after induction of anesthesia demonstrated an intact abdominal aorta and visceral arteries. Just before skin incision, sudden hypotension developed due to pericardial tamponade. The right femoral artery was immediately cannulated, and cardiopulmonary bypass was initiated while the chest was opened for drainage of the tamponade blood. Although this event led to metabolic acidosis, the base excess gradually improved during bypass. At a rectal temperature of 25°C, the ascending aorta was replaced with open anastomosis under circulatory arrest and antegrade cerebral perfusion. During the rewarming period, however, urinary output decreased and metabolic acidosis developed again (base excess decreased from -1.1 mEq/L to -10.3 mEq/L). The TEE demonstrated that blood flow in the visceral arteries was undetectable and that dissection was present in the aorta with the blood flow detectable only within the false lumen. As antegrade flow increased, blood flow became detectable in the visceral arteries and left renal artery. Urinary output gradually increased and the metabolic acidosis disappeared. Although circumferential dissection was apparent in the aorta (Fig 4), there was no apparent dissection in the branch arteries. The subsequent intraoperative and postoperative courses were uneventful. Postoperative computed tomography at 2 weeks postoperatively showed no dissection within the abdominal aorta. The patient’s course has been uneventful up to 4 years after surgery with no aneurysm formation in the aorta.
View larger version (100K):
[in this window]
[in a new window]
|
Fig 4. Transesophageal echocardiogram demonstrating circumferential dissection in the abdominal aorta (AO) after cardiopulmonary bypass in case 11.
|
|
In the remaining 8 cases, aortic dissection was not detected with either intraoperative TEE or postoperative computed tomography, and the postoperative courses during the follow-up period were uneventful. The base excess during cardiopulmonary bypass did not exceed -10 mEq/L in any of these 8 cases.
|
Comment
|
|---|
Our results are summarized as follows. First, new development of aortic dissection after femoral arterial perfusion was detected in 27% of cases in this series in which the abdominal aorta was meticulously examined with intraoperative TEE. Second, this event may be responsible for the unexplained progression of metabolic acidosis that occurs during cardiopulmonary bypass, possibly caused by transient malperfusion of visceral arteries. Third, postoperative assessment can provide false negative information regarding newly developed aortic dissection in the operating room. Fourth, this event may be related to aneurysm formation in the infrarenal and iliac regions during the follow-up period.
Femoral arterial perfusion was likely responsible for new development of aortic dissection because, one, aortic dissection appeared exclusively after femoral arterial perfusion; and two, the intimal flap moved posteriorly, and celiac arterial perfusion was recovered as soon as antegrade perfusion resumed. Retrograde dissection with closure of the false lumen accounts for these findings. A three-dimensionally reconstructed computed tomogram of case 1 is shown in Figure 5. When the internal diameter of the cannula tip was 4 mm, the cross-sectional area of the cannula orifice was calculated to be approximately 0.13 cm2. With a perfusion rate of 2 L/min, the flow velocity at the cannula orifice exceeds 2 m/s. If the external iliac artery curves acutely near the tip of the cannula, as shown in Figure 5, the jet stream ejected out of the cannula could tear the intima or shift the plaque in the external iliac artery, leading to retrograde dissection. The abdominal aorta could be the site for earliest detection of this event through the use of TEE. As it may take some time until the aortic dissection reaches the level of the celiac artery, repeated assessment during cardiopulmonary bypass may be necessary. Such events would have remained undiagnosed if intraoperative TEE had not been used to monitor the abdominal aorta.
View larger version (125K):
[in this window]
[in a new window]
|
Fig 5. Three-dimensionally reconstructed computed tomogram showing how intimal injury occurs by the jet stream from the arterial cannula (arrow) that is inserted into the femoral artery. The flow velocity of jet stream exceeds 2 m/s at the tip of cannula.
|
|
The intimal flap may become pressed against the aortic wall by retrograde blood flow, obstructing the orifice or proximal portion of the branch arteries, leading to malperfusion of the visceral organs. This mechanism may be responsible for unexplained metabolic acidosis that occurs occasionally during cardiopulmonary bypass, because it improves after antegrade perfusion resumes. In this series, marked acidosis (base excess 10 mEq/L) was apparent in 2 of 3 cases with new aortic dissection, but in none of the other 8 cases (p = 0.0545, 
2 test), despite the fact that management during cardiopulmonary bypass was not significantly different.
The present incidence of new dissections (27%) may appear high. It should be emphasized from our results, however, that another drama could be present behind the scenes in these apparently uneventful cases. Transesophageal echocardiography may be a tool capable of revealing such events. Our technique of visualizing the abdominal aorta and visceral branch arteries by means of TEE [5, 6] is beyond the scope of the commonly accepted applications of intraoperative TEE [7]. Some may argue that accuracy of TEE diagnosis of new aortic dissection is questionable. However, the TEE findings shown in the present investigation are not compatible with any category of artifact or other pathologies, based upon the known TEE findings of aortic dissection in the thoracic aorta. In case 1, the accuracy of TEE diagnosis was confirmed by epiaortic echo. The use of sterile echo probes will be helpful for directly visualizing the abdominal aorta and visceral arteries, when the thoracoabdominal approach is taken, as in this case. However, this procedure is less advantageous in cases with a midsternal approach, as in 9 of 11 of the current cases. In case 2, postoperative computed tomography revealed the presence of new aortic dissection. The TEE diagnosis in case 11 may be questionable because aortic dissection was not found in the postoperative computed tomography. However, the intraoperative TEE findings cannot be explained otherwise. On the contrary, the accuracy of computed tomography in diagnosing the presence of aortic dissection may not be perfect. The intimal flap could be pressed against the aortic wall as the antegrade blood flow resumed, while the aorta might appear normal. Theoretically, the visceral perfusion may decrease markedly in deep hypothermia owing to redistribution. Blood flow was clearly detected in the superior mesenteric artery, however, but not in the celiac artery during retrograde perfusion, and became detectable in the celiac artery as soon as antegrade perfusion resumed.
Aneurysmal dilatation appeared in the infrarenal portion of the abdominal aorta or iliac artery, or both, in 2 of the 3 cases with intraoperative development of new aortic dissection. Such events were not encountered in any of the remaining 8 cases. Reduced tolerance to the luminal pressure may be responsible for the aneurysmal change in these cases. This is only speculative, however, based upon our experience with a small patient population, and further investigation is needed to clarify this relationship.
There are several limitations to this study. First, this is a study with very small number of patients, as a consequence of excluding 14 cases in which the abdominal aorta was involved. Nevertheless, findings similar to those obtained in this paper have not yet been reported. Second, the iliac and inferior mesenteric arteries were beyond the scanning area of TEE, and malperfusion in the descending to sigmoid colon was unable to be visualized. Abdominal ultrasonography might be capable of compensating for this limitation. Third, our TEE technique for visualizing the abdominal aorta and visceral branches has not yet been commonly accepted. Manipulation of the TEE probe in the stomach is necessary with this method, and the safety of this technique has not been fully explored in many institutes, although no complications related to this technique have been encountered in our own experiences with hundreds of cases. Fourth, there was a time lag between intraoperative TEE and postoperative assessment. This might be responsible for the disparate results between these two modalities in case 11.
In conclusion, the occurrence of newly developed aortic dissection after femoral arterial perfusion was detected in 27% of all cases by means of TEE in this series. That may be responsible for malperfusion of visceral organs during cardiopulmonary bypass associated with unusual metabolic acidosis and can potentially lead to aneurysm formation in the pelvic region during the midterm follow-up period. Routine use of TEE to examine the visceral arteries is recommended in cases with femoral arterial perfusion.
|
References
|
|---|
- Eugene J, Aronow WS, Stemmer EA. Retrograde aortic dissection during cardiopulmonary bypass Clin Cardiol 1981;4:356-359.[Medline]
- Tanemoto K, Kuinose M, Kanaoka Y. Complications of femoral artery cannulation in aortic arch related operations Nippon Kyobu Geka Gakkai Zasshi 1994;42:306-310.[Medline]
- Pasic M, Schubel J, Bauer M, et al. Cannulation of the right axillary artery for surgery of acute type A aortic dissection Eur J Cardiothorac Surg 2003;24:231-235.[Abstract/Free Full Text]
- Neri E, Massetti M, Capannini G, et al. Axillary artery cannulation in type a aortic dissection operations J Thorac Cardiovasc Surg 1999;118:324-329.[Abstract/Free Full Text]
- Orihashi K, Matsuura Y, Sueda T, et al. Abdominal aorta and visceral arteries visualized by transesophageal echocardiographytechinical considerations. Hiroshima J Med Sci 1997;46:151-157.[Medline]
- Orihashi K, Matsuura Y, Sueda T, et al. Abdominal aorta and visceral arteries visualized with transesophageal echocardiography during operations on the aorta J Thorac Cardiovasc Surg 1998;115:945-947.[Free Full Text]
- Shanewise JS, Cheung AT, Aronson S, et al. ASE/SCA guidelines for performing a comprehensive intraoperative multiplane transesophageal echocardiography examinationrecommendations of the American Society of Echocardiography Council for Intraoperative Echocardiography and the Society of Cardiovascular Anesthesiologists Task Force for Certification in Perioperative Transesophageal Echocardiography. Anesth Analg 1999;89:870-884.[Free Full Text]
Top
Abstract
Introduction
