Anesth Analg 2001;92:335-337
© 2001 International Anesthesia Research Society
CARDIOVASCULAR ANESTHESIA
Sudden Respiratory Arrest Resulting From Brainstem Embolism in a Patient Undergoing Endovascular Abdominal Aortic Aneurysm Repair
Michael Zaugg, MD*,
Mario L. Lachat, MD
,
Thomas Pfammatter, MD
,
Gieri Cathomas, MD§, and
Edith R. Schmid, MD*
*Division of Cardiovascular Anesthesiology and the Departments of Cardiovascular Surgery, Radiology, and §Pathology, University Hospital Zurich, Zurich, Switzerland
Address correspondence and reprint requests to Michael Zaugg, MD, Institute of Anesthesiology, University Hospital Zurich, Rämistrasse 100, 8091 Zurich, Switzerland. Address e-mail to michael.zaugg{at}ifa.usz.ch
 |
Abstract
|
|---|
Implications: Surgery on the aorta is a great challenge for the anesthesiologist, even with newly developed and less invasive stent-graft procedures. The case of a fatal cerebral embolism during endovascular repair of an abdominal aortic aneurysm is reported, and the potential mechanisms underlying this unexpected complication are discussed.
|
Introduction
|
|---|
Stent-graft procedures for repair of abdominal aortic aneurysm (AAA) are increasingly used because of reduced perioperative stress response, improved intraoperative hemodynamic stability, and decreased incidence of embolic complications (1,2). Earlier discharge and reduced costs have been reported with this less invasive technique, as well (3). In contrast to the conventional open AAA repair, endografting can be performed either under regional anesthesia, or using IV analgesia supplemented with local anesthetic infiltration at the sites of surgical access. In this report, we present the case of a fatal intraoperative cerebral insult in a patient undergoing endovascular AAA repair.
|
Case Report
|
|---|
A 72-yr-old male was admitted for coronary revascularization and for repair of a 2 x 5 cm infrarenal AAA. His medical history included previous myocardial infarction, mild renal insufficiency (creatinine 121 µM, normal range: 70–105 µM), and pulmonary embolism as a result of a deep venous thrombosis of his right leg approximately 1 yr ago. Medications included pravastatin, aspirin, nifedipine, sotalol, enalapril, furosemide, nitroglycerin, and phenprocoumon until 1 wk before coronary revascularization, when phenprocoumon was replaced by heparin because of newly developed unstable angina. Preoperative transthoracic echocardiography revealed anterolateral hypokinesia with an ejection fraction of 60%. Duplex sonography of the carotid arteries was unremarkable. The patient first underwent uneventful coronary artery bypass surgery with revascularization of the left descending and left circumflex coronary arteries. Marked atheromatous plaques were observed by intraoperative transesophageal echocardiography (TEE) in the aortic arch as well as descending thoracic aorta. Postoperative TEE revealed decreased anterolateral hypokinesia and a slightly improved ejection fraction of 70%. All cardiovascular and antithrombotic medications were immediately resumed after successful coronary revascularization.
For prognostic reasons, it was decided to repair the progressively increasing and partially thrombosed infrarenal AAA during the same hospital stay by means of endovascular technique. The procedure was performed in the completely recovered patient under monitored anesthesia care with IV analgesia using remifentanil (0.01–0.1 µg · kg-1 · min-1) and local anesthetic infiltration with lidocaine 0.5% (total administered dose 2 mg · kg-1 body weight) into the right and left groins. Activated clotting time was increased from 120 s to 410 s by IV administration of 5000 U of heparin. Standard monitoring included 12-lead electrocardiogram, pulse oximetry (right index finger), invasive blood pressure (left radial artery), and central venous pressure (right internal jugular vein) measurements. Mean arterial blood pressure was adjusted to 60 mm Hg using continuous nitroglycerin infusion of 0.5–2 µg · kg-1 · min-1. Heart rate ranged between 60–70 beats per min in the ß-blocked patient with no further treatment. Under this anesthetic management, the patient was completely free of pain, responsive, and breathing sufficiently with oxygen saturation more than 95%. Ninety minutes after starting preparation of the operative field, the endovascular prosthesis (Excluder 34 x 12; Prograft Medical Inc., Sunnyvale, CA) was deployed via the right common femoral artery, and positioned by means of balloon inflation with its proximal end below the lower renal artery. At this time, the patient started gasping for breath and finally stopped breathing within seconds. No signs of seizures were present. Mean arterial blood pressure remained unchanged, whereas heart rate slightly decreased to 55 bpm and oxygen saturation decreased to 83%. The patient was immediately ventilated by mask with 100% oxygen, and the trachea was intubated, facilitated by 12 mg etomidate and 100 mg rocuronium. Anesthesia was maintained with 100% oxygen and propofol infusion. Clinical evaluation revealed bilateral dilated pupils, which were not reactive to light. Although blood gas analysis from the left radial artery demonstrated an oxygen saturation of 100% with a PO2 of 165 mm Hg, pulse oximetry, measured on the right index finger, remained less than 90%, and only recovered gradually over the next hour to 95%. Conversely, oxygen saturation on the left and right lobule of the ear was 100%, in accordance with blood gas analysis. No cutaneous manifestations, such as livedo reticularis or "purple toe syndrome," characteristic of cholesterol embolism, were present. Immediately performed intraoperative duplex sonography of the carotid arteries and transcranial Doppler sonography did not reveal any occlusion or uncommon flow patterns. TEE showed well-preserved cardiac function and no change in the pattern of atheromatous plaques present in the aorta. There was no thrombus formation, nor was there a patent foramen ovale. Because of continued hemodynamic stability, it was decided to complete the proper positioning of the prosthesis.
Postoperative computed tomography scan, initially without contrast medium enhancement because of preexisting increased creatinine serum values, demonstrated a massive ischemic brainstem insult with multiple foci including pons, cerebellum, and thalamus. Clinical evaluation after cessation of sedation revealed bilateral positive Babinski’s sign, generalized convulsions to pain stimulation, and persisting dilated pupils without reaction to light. Vestibulo-ocular reflex was absent. The diagnosis was made of a massive ischemic brainstem insult, most likely resulting from a thromboembolic event. The patient was fully heparinized and ventilated with 50% oxygen. Laboratory data revealed increased total creatine kinase (maximum 520 U/L, normal range <270 U/L) without significant increase in MB isoenzyme fraction or cardiac troponin T). A blood sample immediately drawn after the acute event did not show toxic lidocaine serum levels (2.2 µM, normal range: 6.4–21.3 µM). A repeated computed tomographic scan, enhanced with contrast medium one day after the event, clearly showed the occlusion of the basilar artery as the cause of the vast ischemic infarctions of the brainstem ( Fig. 1A). The patient died three days after the endovascular procedure.
View larger version (86K):
[in this window]
[in a new window]
|
Figure 1. A, contrast medium enhanced computed tomography scan of the head at the base of the brain with ischemic lesions, edema, and missing depiction of the basilar artery (arrowhead) indicating total occlusion of this vessel. B, proper placement of the endovascular stent-graft in the partially thrombosed abdominal aortic aneurysm. The self-expandable stent-graft requires several short episodes of total aortic occlusion, each lasting 20–30 s. C, massive plaque formation in the aortic arch and descending thoracic aorta.
|
|
Autopsy demonstrated open coronary artery bypass grafts and confirmed proper placement of the endovascular prosthesis in the partially thrombosed AAA (Fig. 1B). There was no fresh deep venous or cardiac thrombosis, nor was there recent pulmonary embolism. Extensive atheromatous plaques were observed in the entire aorta (Fig. 1C). Massive destruction and colliquation of the brainstem was present and consistent with total occlusion of the basilar artery. No cholesterol emboli were detected on peripheral brain sections. Further evaluation of basal brain vessels with respect to localization of embolic material was not feasible because of extensive autolytic destruction of the tissue. No dissections along the blood vessels supplying the brain were observed. Residual thromboembolic material was detected in the right radial artery.
|
Discussion
|
|---|
In contrast to open repair, complications with the less invasive endovascular repair of AAA tend to be local. However, major acute complications such as aneurysm rupture, abdominal, pelvic, or lower extremity ischemia, malposition and migration of stents, and distal embolization of plaques or thrombi have been described (4,5). We report for the first time on a case of fatal cerebral embolism during endovascular grafting for AAA. The massive brainstem embolism, which manifested by loss of consciousness and sudden respiratory arrest, was rapidly diagnosed in this otherwise responsive and only moderately sedated patient. Local anesthetic toxicity, and overdose with sedatives or analgesics could be easily excluded.
Although the massive necrosis and softening of the brainstem did not allow precise localization of embolic material, the selective extent of the ischemic damage to basal brain structures at autopsy, as well as the total occlusion of the basilar artery as evidenced by contrast medium enhanced computed tomographic scans, are highly suggestive for embolism into the basilar artery. Likewise, the observed sustained discrepancy between oxygen saturation, measured by pulse oximetry on the right index finger, and oxygen saturation measured via blood gas analysis, further supports the occurrence of massive thromboembolism into the right subclavian artery with subsequent occlusion of the basilar artery. In fact, residual embolic material was subsequently detected in the right radial artery at autopsy. Although the exact composition of the embolic material was not determined microscopically, it seems very likely that the embolic material derived from large atheromatous plaques or thrombi of the aortic wall, which were released by surgical manipulations. This is further supported by the fact that the embolic event coincided with increased surgical activity, i.e., advancement of guiding wires, device deployment, balloon inflation, and vigorous flushing of catheters with large volumes of contrast medium. Distal embolization of thrombotic material is not uncommon in the setting of endovascular AAA repair. However, embolization to a large branch of the aortic arch, has not been observed during this procedure.
Several potential mechanisms underlying the embolic event need to be considered. Paradoxical embolism via a patent foramen ovale was excluded by TEE and at autopsy. Advancement of guidewires up to the aortic arch with subsequent damage and release of embolic material may have occurred. Indeed, intracardiac guidewire placement with subsequent severe aortic regurgitation was recently reported (6). However, no freshly damaged atheromatous plaques were observed in the aortic arch at autopsy. Another putative mechanism of embolism may have been the retrograde expulsion of thrombotic material at the time of temporary balloon inflation. Retrograde migration of a large thrombus from the proximal left anterior descending artery to the left circumflex artery was previously reported during coronary angioplasty (7). The underlying mechanism is thought to be the retrograde expulsion of thrombotic material by the deploying balloon. Alternatively, dislodgement of thrombotic material from the partially thrombosed aneurysm may have occurred by forceful injection of large volumes of contrast medium. Also in this case, the thrombus is postulated to travel retrogradely against the arterial pressure up to the aortic arch into one of the major cerebral arteries. The potential and serious hazard of forceful flushing of arterial cannulas was previously recognized (8). Notably, occlusion of the basilar artery with sudden loss of consciousness and subsequent death as a result of brainstem infarction, similar to the presented case, was described after the injection of a small volume of isotonic saline into a clotted Scribner shunt (9).
Various strategies to avoid stent migration, proximal hypertension, and undue strain on myocardium and aortic wall, including vena cava occlusion with a balloon (10), electrically induced ventricular fibrillation (11), and adenosine-induced temporary asystole (12) are currently used during endovascular repair of AAA. These techniques lead to nearly complete stop of aortic flow during the critical episodes of balloon inflation. In the case presented, nitroglycerin was successfully used to control blood pressure and avoid myocardial damage (13). So far, however, it is not clear whether these hemodynamic alterations prevent or rather precipitate the occurrence of cerebral embolization.
In conclusion, cerebral embolization may occur during the critical episodes of endovascular repair of AAA, which can be easily detected in a sedated but conscious patient. The potential mechanisms underlying this complication include antegrade as well as retrograde migration of emboli released by surgical manipulations in the aorta. Importantly, forceful large-volume flushing of aortic cannulas, as a potential mechanism of retrograde embolization, should be meticulously avoided, particularly in patients with extensive luminal pathology of the aorta.
|
References
|
|---|
-
Blum U, Voshage G, Lammer J, et al. Endoluminal stent-grafts for infrarenal abdominal aortic aneurysm. N Engl J Med 1997; 336: 13–20.[Abstract/Free Full Text]
-
Elmarasy NM, Soong CV, Walker SR, et al. Sigmoid ischemia and the inflammatory response following endovascular abdominal aortic aneurysm repair. J Endovasc Ther 2000; 7: 21–30.[ISI][Medline]
-
Brewster DC, Geller SC, Kaufman JA, et al. Initial experience with endovascular aneurysm repair: comparison of early results with outcome of conventional open repair. J Vasc Surg 1998; 27: 992–1005.[ISI][Medline]
-
Moskowitz DM, Kahn RA, Marin ML, Hollier LH. Intraoperative rupture of an abdominal aortic aneurysm during an endovascular stent-graft procedure. Can J Anaesth 1999; 46: 887–90.[Abstract/Free Full Text]
-
Bertrand M, Godet G, Fléron M-H, et al. Lumbar muscle rhabdomyolysis after abdominal aortic surgery. Anest Analg 1997; 85: 11–5.[Abstract]
-
Sutton DC, Rother A. Endoluminal abdominal aortic aneurysm repair complicated by intracardiac guidewire placement and massive transfusion. Anesth Analg 2000; 91: 89–91.[Free Full Text]
-
Desmet WJ, Dens J, Piessens J. "Back-squeezing" of the clot: an unusual complication of primary coronary angioplasty. Cathet Cardiovasc Diagn 1997; 42: 64–7.[ISI][Medline]
-
Lowenstein E, Little JW, Har Lo W. Prevention of cerebral embolisation from flushing radial-artery cannulas. N Engl J Med 1971; 285: 1414–5.
-
Gaan D, Brewis RAL, Mallick NP, et al. Cerebral damage from declotting Scribner shunts. Lancet 1969; 2: 77–9.[ISI][Medline]
-
Nishikimi N, Usui A, Ishiguchi T, et al. Vena cava occlusion with balloon to control blood pressure during deployment of transluminally placed endovascular graft. Am J Surg 1998; 176: 233–4.[ISI][Medline]
-
Kahn RA, Marin ML, Hollier L, et al. Induction of ventricular fibrillation to facilitate endovascular stent graft repair of thoracic aortic aneurysm. Anesthesiology 1998; 88: 534–6.[ISI][Medline]
-
Dorros G, Cohn JM. Adenosine-induced transient cardiac asystole enhances precise deployment of stent-grafts in the thoracic or abdominal aorta. J Endovasc Surg 1996; 3: 270–2.[ISI][Medline]
-
Bernard E, Schmid ER, Lachat ML, Germann RC. Nitroglycerin to control blood pressure during endovascular stent-grafting of descending thoracic aortic aneurysm. J Vasc Surg 2000; 31: 790–3.[ISI][Medline]
Accepted for publication October 26, 2000.
This article has been cited by other articles:
|
|
|
|
 
S. Tesoro, D. Mezzetti, L. Marchesini, and V. A. Peduto
Clonidine Treatment for Agitation in Children After Sevoflurane Anesthesia
Anesth. Analg.,
December 1, 2005;
101(6):
1619 - 1622.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
