点击显示 收起
1 From the Departments of Interventional Neuroradiology (R.C., E.H., J.P.S.M., A.A., C.M., J.J.M.) and Neurosurgery (G.L.), Hôpital Lariboisière, 1 rue Ambroise Paré, 75475 Paris Cedex 10, France. Received February 8, 2001; revision requested March 26; revision received June 26; accepted July 16.
| ABSTRACT |
|---|
| Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
MATERIALS AND METHODS: Clinical and radiologic data of 18 MAs treated with EVT were retrospectively reviewed. There were 14 patients (11 men, three women), ranging in age from 28 to 64 (mean age, 44 years). All patients had endocarditis and positive blood culture. The aneurysms were located within the distal cerebral circulation (n = 13) or in the circle of Willis (n = 5). There were 12 ruptured aneurysms and six unruptured aneurysms. Distal or fusiform aneurysms were treated by means of parent vessel occlusion. Proximal saccular aneurysms were selectively treated.
RESULTS: Endovascular treatment was successful for all aneurysms. No aneurysm bled after embolization during clinical follow-up. Follow-up angiograms obtained in 11 of 14 patients 6 months to 2 years after the procedures showed stable occlusions. Transient complications occurred in two cases, with worsening of hemiparesis and quadrantanopia. Five patients underwent surgical cardiac valve replacement within 1 week of EVT without neurologic complications. The late clinical outcome was normal neurologic status (n = 9) or permanent disability that was related to the initial stroke (n = 5).
CONCLUSION: EVT is a reliable and safe technique that should be considered at the time of diagnosis of cerebral mycotic aneurysms.
Index terms: Aneurysm, cerebral, 178.733 • Aneurysm, mycotic, 178.733 • Aneurysm, therapy, 178.1264 • Arteries, therapeutic embolization, 178.1264 • Endocarditis, 512.21
| INTRODUCTION |
|---|
| Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
| MATERIALS AND METHODS |
|---|
| Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
Patient Population
Fourteen consecutive patients (three women, 11 men; age range, 28–64 years; mean age, 44 years) were referred to our hospital between 1991 and 1999 for EVT of a MA, for which informed consent was obtained. EVT for cerebral aneurysms is a clinically accepted, routine procedure at our institution; therefore, our ethics committee did not require its approval. Our ethics committee also does not require its approval or informed patient consent for the retrospective review of clinical and radiologic data. A report of the first three patients included in this study has been published previously (6).
All patients had endocarditis with valve vegetations on cardiac ultrasound images. Blood cultures were positive in all patients. Bacterial cultures revealed infection with Streptococcus (n = 8) or Staphylococcus (n = 6) organisms.
Of 14 patients, 11 had acute neurologic symptoms (eight patients had hemiparesis or hemiplegia, two had meningism, and one had hemianopia). Three patients were free of neurologic symptoms.
Computed tomographic (CT) scans in the 11 patients with acute neurologic symptoms revealed the following isolated or combined lesions: parenchymal hematoma (n = 8), ischemic stroke (n = 4), and subarachnoid hemorrhage (n = 3). Clinically silent MAs were suspected in the other three patients because of the presence of hyperattenuating, punctiform areas on CT scans obtained after the administration of contrast material.
Magnetic resonance (MR) angiography was performed in one patient and failed to demonstrate a small distal MA.
Diagnosis was achieved at angiography in all cases. Eighteen MAs were treated in 14 patients. Eleven patients had one MA, two patients had two MAs, and one patient had three MAs. There were 13 distal MAs and five proximal MAs. The aneurysmal sac was either fusiform (n = 14) or saccular (n = 4) and ranged in size (ie, widest diameter) from 1 to 11 mm (mean size ± SD, 4 mm ± 3). The combination of clinical data, CT scan features, and angiographic findings were used to help determine which aneurysms had ruptured. Eight MAs ruptured once, four ruptured on two occasions, and six remained unruptured.
Evolution of aneurysm morphology was observed on angiograms of eight MAs. In four cases a MA newly or first appeared on a second angiogram, in three cases the initial aneurysm increased in size, and in one case the initial aneurysm decreased in size. These changes occurred within 1 week to 3 months of the start of antibiotic treatment.
Seven MAs ruptured during antibiotic treatment. No previous angiogram had been obtained in three of these seven MAs, a previous angiogram was normal in three MAs, and a previous angiogram had demonstrated the remaining MA.
In addition to the 18 MAs initially included in our study, a 19th MA was found incidentally in one patient who had a ruptured mycotic aneurysm; this second, unruptured aneurysm resolved completely with antibiotic treatment within 2 months and was excluded from this study.
Associated extracranial MAs were found in five patients in the arteries of the lower limbs (n = 3), in the mesenteric arteries (n = 2), and in the coronary artery (n = 1). Two of these six aneurysms resolved with antibiotic treatment, three were successfully treated with an endovascular procedure, and one required surgical repair.
Treatment of MA
All patients were treated with antibiotics of the appropriate sensitivities.
Three patients who had both a MA and a compressive hematoma were treated with surgery and were not included in this study. The timing of EVT depended on whether the aneurysm had ruptured. In our initial six patients, ruptured aneurysms were treated immediately, whereas unruptured aneurysms were not treated until the end of a course of antibiotics. In the later eight patients in our series, however, we treated unruptured aneurysms with EVT at the time of diagnosis.
Technique
Patient informed consent was obtained for the procedures. EVT procedures were performed after the patients had been given general anesthesia or sedation. Sedation was used in cooperating patients when occlusion of a parent vessel was required and/or when a Wada test was considered. A routine coaxial technique with the femoral arterial route was used. A 6-F guiding catheter was inserted at the cervical level with continuous flushing with normal saline. The patients underwent anticoagulation with intravenous heparin to obtain an activated partial thromboplastin time two to three times baseline.
EVT of Distal Aneurysms
A flow-guided microcatheter (Magic, Balt, Montmorency, France; or Spinnaker, Target Therapeutics, Fremont, Calif) or an over-the-wire microcatheter (Excel 14, Target Therapeutics; Tracker 18, Target Therapeutics; or Rapid Transit, Cordis, Miami, Fla) was used. The microcatheter was positioned as close as possible to the MA. If the MA could not be reached with the microcatheter, the parent artery was occluded. If the MA could be reached with the microcatheter, the occlusion included the parent artery and the aneurysmal sac. A Wada test with selective intraarterial injection of 25 mg amobarbital (Amytal; Lilly, Saint-Cloud, France) was performed in patients with aneurysms that were located in highly functional territories when occlusion of a parent vessel was required.
EVT of Proximal Aneurysms
Selective treatment of proximal saccular MAs was performed with a technique similar to that used in the treatment of berry aneurysms.
Embolization Material
For parent vessel occlusion, we used cyanoacrylate (Histoacryl; B Braun, Melsungen, Germany), which was mixed with iodized oil (Lipiodol Ultra Fluide; Laboratoire Guerbet, Paris, France) in ratios varying from 1:1 to 1:3. The following materials were used in one procedure each: autologous clot, polyvinyl alcohol microparticles (Contour; Target Therapeutics), and a straight coil (Target Therapeutics). Selective aneurysmal occlusion was performed with Guglielmi detachable coils (Target Therapeutics) when they were available. In one of the first aneurysms encountered in our study, a detachable latex balloon (Goldvalve no. 17; Nycomed Amersham, Paris, France) and fibered coils (Hilal coils; Cook, Bloomington, Ind) were used to achieve occlusion. Autologous clots, balloons, and fibered coils were used only in our early experience, when more controllable embolic materials were not available.
Follow-up
All patients were evaluated clinically and underwent cerebral angiography of four vessels within 1 week of EVT (R.C., E.H., A.A.). Eleven patients also underwent clinical and angiographic follow-up between 6 months and 2 years after EVT (R.C., J.P.S.M.). Three patients were followed up with clinical examinations alone (R.C.) 2 years after EVT. Clinical status was evaluated with the Rankin scale, and the stability of aneurysm occlusion was evaluated at angiography.
| RESULTS |
|---|
| Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
|
Of 13 distal MAs, nine were treated with sealing of both the parent vessel and the aneurysm with cyanoacrylate (Fig 1). In the treatment of the other four distal MAs, just the parent artery was occluded (Fig 2) with either cyanoacrylate (n = 1), autologous clot (n = 1), polyvinyl alcohol microparticles (n = 1), or a combination of a straight coil and cyanoacrylate (n = 1). A Wada test was performed in four patients, which led to distal repositioning of the tip of the microcatheter before EVT in one patient who developed a transient deficit after the test.
|
|
|
|
|
|
|
|
|
|
|
The late follow-up angiograms showed stable aneurysm occlusion.
There were no clinical changes after treatment of 16 MAs. Transient complications occurred in two patients with distal MAs. In one patient with a parietal hematoma, occlusion of a Rolandic artery led to a slight worsening of preexisting hemiparesis. The hemiparesis resolved completely within 1 year. In the second patient, occlusion of the parent vessel of a MA induced a transient quadrantanopia that resolved spontaneously within 48 hours. These two complications occurred in one of 13 ruptured MAs and in one of five unruptured MAs. Hence, the morbidity of EVT procedures was not significantly higher in patients with unruptured MAs than in patients with ruptured MAs.
Five patients underwent cardiac surgery within 1 week of EVT without cerebral complications.
Overall, measures of the late clinical outcome of the patients revealed 0% mortality and 35% morbidity. Of the five patients who experienced morbidity, one was assigned a Rankin score of 2, two were assigned a Rankin score of 3, and two were assigned a Rankin score of 4. In all five patients, disability was due to the initial hemorrhage or ischemic stroke.
| DISCUSSION |
|---|
| Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|
EVT Technique for Distal MAs
The decision to use a coil or cyanoacrylate as an occlusive device depends on the location of the MA. If the aneurysm can be reached with a microcatheter, cyanoacrylate is the logical choice because it permits simultaneous sealing of the aneurysm and the parent vessel. A rapid burst of increased pressure has been predicted with the use of cyanoacrylate (11), but in our experience, no aneurysms ruptured after careful injection. A coil may also be used to achieve occlusion of the parent vessel (11). We did use a coil in one patient, but the advantage of using Guglielmi detachable coils (ie, the ability to retrieve and reposition a coil) must be weighed against the elevated risk of perforating the inflamed and fragile aneurysmal wall. Furthermore, a Guglielmi detachable coil requires a larger microcatheter.
If the aneurysm cannot be reached with the microcatheter, the occlusion device must be chosen on the basis of the risk involved in using each device. The use of a coil may lead to distal refilling of the MA through collateral vessels, as occurred in patient 7, but injection of diluted cyanoacrylate may induce an ischemic stroke when a large arterial segment is occluded. In this situation, functional testing of the patient with amobarbital may help determine if injection of cyanoacrylate is possible or if the use of a coil is preferable.
In general, polyvinyl alcohol microparticles should be used to treat only very distal aneurysms, in which a distal occlusion of the arterial tree must be achieved and in which the microcatheter cannot be positioned close to the aneurysm.
EVT Technique for Proximal MAs
Balloons are no longer used for selective aneurysmal treatment and have been replaced by detachable coils (15). Selective treatment of saccular aneurysms can be achieved with coils, even if the aneurysms are MAs.
Results of EVT
Local infection.—Theoretically, the introduction of foreign material into an infected area may result in prolonged infection and abscess formation. Such complications did not occur in our experience, although results of blood cultures did not influence the timing of the EVT procedures. However, EVT was always performed while patients were receiving antibiotic treatment, and it may have been the concurrent use of antibiotics that prevented this complication.
Ischemic complications.—The low rate of ischemic complications after parent artery occlusion can be explained by the pathophysiology of MAs. The first event in the creation of a MA is the migration of a septic embolus into the cerebral circulation until it lodges in a vessel (16). This embolus is sometimes visible at angiography (17,18). Local spread of the bacterial agent leads to vessel wall necrosis and the development of a false aneurysm (19). At the same time, arterial occlusion may induce an ischemic stroke, which will either be clinically relevant or remain silent. Ischemic stroke is sometimes avoided when retrograde filling of the occluded artery through leptomeningeal anastomoses occurs. Therefore, therapeutic occlusion of a vessel at the site of a MA is unlikely to induce a stroke, because a stroke would either already have occurred or would have been avoided by the presence of collateral vessels. Both cases of ischemic complications that we encountered may be explained by occlusion of normal vessels.
Late outcome.—Late outcomes in this series were good, with no mortality and a disability rate of 35% that was related to initial stroke. In series of more than five patients with MAs, mortality ranges from 20% (20) to 83% (2) after treatment with antibiotics and from 7% (21) to 61% (22) after surgery and administration of antibiotics. It is likely that some bias existed in these studies; surgical patients may have been treated primarily for evacuation of hematomas with aneurysm excision as an adjunct. These patients may have more severe initial neurologic deficits than those treated with EVT, and the late clinical outcome in these patients may be poorer than in patients without compressive hematoma who are treated with EVT. In our series, patients with compressive hematoma were treated with surgery. On the other hand, nonsurgical patients may also have a poor outcome because they may not have been accepted for surgery because of their poor condition.
EVT versus Surgery
A distal aneurysm can sometimes be missed with a surgical approach (21), so stereotaxy has been recommended (23,24). The risks of surgery are greater for patients with large, multiple, or ruptured MAs and for patients with MAs located in highly functional areas (3,22, 25). Delayed surgery has sometimes been advocated in the hope that fibrosis of the aneurysmal wall may reduce the perioperative risk of rupture (22,26); however, bleeding may occur during this waiting period because the risk of rupture is higher during the acute phase (3).
EVT is less invasive than surgery and may be performed with sedation, which permits functional testing of the patient before vessel occlusion (27). In our experience, unlike the results of surgery, the results of EVT are not influenced by the timing of the procedure, by previous rupture, or by the number and location of MAs. Furthermore, early aneurysmal occlusion enables optimal management of endocarditis and rapid cardiac surgery because a mechanical-valve prosthesis (which requires anticoagulation) can be used after cure of the MA (because anticoagulation cannot be used in the presence of a MA) (28). If a MA is still present, however, only a bioprosthesis (which does not require anticoagulation) can be used.
EVT versus Medical Treatment
Conservative medical treatment with antibiotics alone and without selective aneurysmal occlusion has been proposed for patients with MAs (4). Medical treatment is necessary for the underlying endocarditis and reduces the MA-related hemorrhagic risk (18). Some MAs may even resolve with antibiotic treatment (4,29,30). It can therefore be argued that EVT should be performed only after a full course of treatment with antibiotics. However, the outcome for patients with MAs after antibiotic treatment remains highly variable. All types of aneurysmal evolution were observed in our series during antibiotic treatment: reduction of the aneurysm, enlargement of the aneurysm (Figs 2, 3), formation of new aneurysms, and rupture of the original aneurysm. In a study of 14 patients with 18 MAs treated with conservative measures followed by surgery, antibiotics led to resolution of the aneurysm in four patients, but one patient died of hemorrhage before surgery (21). The different outcomes of MAs during and after antibiotic treatment have been studied in an extensive review of the literature (1). Complete aneurysmal regression was found in 20 reports, aneurysmal rupture was found in 17, aneurysmal enlargement was found in five, and de novo aneurysmal appearance during antibiotic treatment was found in five. The uncertainty of aneurysmal evolution is underlined by the independent evolution of each MA in our patients with multiple MAs. In patient 5, two MAs in the mesenteric arteries resolved during antibiotic treatment, while a cerebral MA ruptured twice and another cerebral MA enlarged.
Identification of predictive factors for aneurysmal rupture would be useful, but, to our knowledge, none have been recognized (4). Slow filling of MAs at angiography and stability of the aneurysm at follow-up angiography have been suggested as factors that indicate a reduced risk of rupture (22). However, these observations were made on the basis of experience with only a few patients and, to our knowledge, have not been confirmed.
Considering the variability of outcomes with antibiotic treatment and the good results of EVT, we recommend that EVT be considered at the time of diagnosis for all patients with cerebral MAs.
Ruptured versus Unruptured MAs
Unlike treatment of congenital aneurysms, our treatment of MAs is not influenced by a history of rupture. Indeed, to our knowledge, there is no support in the literature for a difference in the risk of future rupture between ruptured and unruptured MAs. Patients with ruptured MAs have higher mortality than patients with unruptured MAs (1), but this may be the consequence of hemorrhage rather than rebleeding. It has also been suggested that the risk of rupture is higher in the acute phase (3), but this does not imply a higher rate of hemorrhage. Because the pathophysiology of MAs implies vessel wall necrosis, there is a risk of hemorrhage as soon as a MA appears. As long as the bleeding risk of MAs is unknown, and given the good results of EVT, this treatment should be considered at the time of diagnosis for patients with ruptured and unruptured MAs.
Screening
The incidence of MAs remains uncertain, although they are currently believed to occur in 1%–10% of patients with endocarditis (1). However, this incidence is probably an underestimation because some MAs are asymptomatic and are found incidentally. Additionally, some MAs may be missed at angiography due to the great variability in their timing (14). Some MAs may also be missed at histopathologic analysis, where they can be difficult to find (31–33). Therefore, MAs should be investigated, perhaps with contrast material–enhanced CT and MR imaging, to determine their true incidence, to enable a better estimation of their bleeding risk, and perhaps to enable prediction of those MAs that are at increased risk of rupture. Systematic screening for MAs with contrast-enhanced CT has been recommended in all patients with endocarditis (5), even in the absence of neurologic symptoms (22,34). Repeated angiography should be performed when neurologic symptoms are present (5) or if a clinically silent ischemic lesion is found on a CT scan. The ability of CT angiography or MR angiography to depict MAs is unknown and must be evaluated; it is likely to be low because many MAs are small, peripheral, slow-flow aneurysms.
Conclusion
EVT is a safe and effective technique that seems preferable to surgery and should be considered at the time of diagnosis for all patients with cerebral MAs. In our experience, EVT improved the outcome for patients with cerebral MAs.
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
| Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES |
|---|