| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Diagnostic Radiology, University of Bern, Inselspital, Freiburgstrasse 20, CH 3010 Bern, Switzerland. Received May 22, 2001; revision requested July 9; revision received September 26; accepted November 12. Address correspondence to H.P.D. (e-mail: hans-peter.dinkel@insel.ch).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Six consecutive patients (three men, three women; mean age, 46 years, age range, 18–74 years) underwent arterial embolization of nine PAVMs with superselective catheterization with a 3-F coaxial catheter system and embolization with 0.018-inch microcoils. The PAVMs varied in size from 1 to 6 cm (mean, 2.5 cm). Five of the lesions were also treated by filling the venous sac with GDCs. Success and outcome were evaluated by means of a review of patient records, angiographic findings, and telephone interview results.
RESULTS: Complete primary occlusion was achieved in eight of nine lesions; repeat embolization resulted in successful occlusion of one lesion. The superselective technique enabled successful embolization in one patient after a previous procedure performed with a 0.035-inch (Gianturco) coil had failed. Filling of the venous sac was performed in the presence of dilated draining veins and enabled successful occlusion of the feeding artery with microcoils in all cases. There were no complications.
CONCLUSION: Superselective embolization with microcatheters allowed easy catheterization and safe coil deployment. Filling of the venous sac reliably prevented systemic migration of GDCs in PAVMs with a large venous component.
© RSNA, 2002
Index terms: Arteriovenous malformations, pulmonary, 94.149 • Arteriovenous malformations, therapeutic embolization, 94.1264 • Lung, vascular disease, 60.1494
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Although a number of studies have focused on embolotherapy of PAVMs, most have originated from only three institutions (Yale University School of Medicine, New Haven, Conn; Hammersmith Hospital, London, England; and Hospital Calmette University Center, Lille, France) (4,9,11,13–23) and have almost exclusively involved direct embolization with 0.035-inch coils or balloons. The purpose of our study was to assess the value of superselective embolotherapy of PAVMs with coaxial microcatheters and 0.018-inch microcoils. A further purpose was to evaluate our experience with the technique of filling the venous sac with Guglielmi detachable coils (GDCs); this procedure was also performed in half of the patients discussed here.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
Angiographic and Embolization Techniques
Arterial pulmonary angiograms were obtained with a 5.5-F Grollmann pulmonary pigtail catheter (Cook, Bloomington, Mass) inserted via the femoral vein with a 7-F introducer. Selective angiograms were then obtained with a 5-F cobra catheter (Cook). The feeding artery of the PAVM was identified and catheterized superselectively in all patients with a coaxially placed 3-F catheter (Tracker-18 two-tip; Boston Scientific/Target Therapeutics, Natick, Mass) that was advanced through the cobra catheter by means of a 0.016-inch guide wire (Taper-16; Boston Scientific/Target Therapeutics).
All procedures were performed while patients were receiving drugs for local anesthesia without sedation. The patients were monitored with electrocardiography and noninvasive pressure measurements by an anesthesiologic team consisting of a nurse acting under the supervision of an anesthesiologist. Arterial oxygen saturation was measured noninvasively with digital pulse oximetry. Embolization was achieved with fibered 0.018-inch platinum microcoils introduced by means of a 0.016-inch coil pusher (Coil Pusher-16; Boston Scientific/Target Therapeutics).
The microcoils used were VortX-18 vascular occlusion coils (Boston Scientific/Target Therapeutics) of 2–6 mm in unconstrained diameter and 22–85 mm in stretched length or Tracker-18 BOD coils (Boston Scientific/Target Therapeutics) of 5–10 mm in unconstrained diameter and 14–28 mm in stretched length. We used coils of nominal diameters that were 10%–20% larger than the estimated diameter of the feeder artery (ie, oversizing).
To prevent potential systemic migration of the smaller coils used to occlude the feeder artery, we occluded the venous sac of five PAVMs in three patients. These five PAVMs had a large venous sac combined with a dilated draining vessel. The venous sac was catheterized superselectively with a coaxial microcatheter and was densely filled with 0.018-inch GDCs (GDC Standard Coils; Boston Scientific/Target Therapeutics) of 10–20 mm in unconstrained diameter and 30 cm in stretched length; the feeding artery was then occluded with 0.018-inch microcoils. However, PAVMs with large draining veins were not always treated in this way; in complex PAVMs and in PAVMs with tapering feeding arteries in which the chances of coil migration were considered low, filling of the venous sac was not performed. The number and the type of the coils used in each procedure were obtained from our radiologic records, which include a detailed list of materials used. The Figure illustrates our technique.
|
|
|
|
Follow-up pulmonary angiography was performed 1 day up to 6 weeks after the embolization procedure to rule out residual perfusion of the PAVMs. Clinical follow-up was conducted by the referring pulmonologists and family practitioners. Clinical success was defined as the documented complete disappearance of any symptoms of dyspnea and the absence of a new onset of neurologic symptoms potentially attributable to central embolism. Major complications were defined as any clinical or imaging signs of systemic embolization, such as limb ischemia, bowel infarction, stroke, transient ischemic attacks, and visual disorders. Iliac and common femoral deep venous thrombosis, pulmonary embolism, and bleeding were also defined as major complications. Minor complications were defined as pleurisy, segmental or subsegmental lung infarction, superficial femoral or calf thrombosis, and cardiopulmonary changes that required medical intervention. We did not include extrasystoles in this category because they are an expected side effect of the passage of catheters and guide wires through the right atrium.
Assessment of PAVM Morphology
All angiograms were reviewed by H.P.D., who took compass measurements of the sizes of the supplying arteries and veins as they appeared on overhead projection images. Diameters of the relevant structures were tabulated at the point of their greatest diameter on selective pulmonary angiograms obtained in the perpendicular projection; the size of each structure was then calculated according to the following equation: X = (a/b) x (5 mm/
), where a is the greatest diameter of the structure and b is the projected diameter of the 5-F selective catheter on the same image. All resulting measurements were rounded to full millimeters. The size of the PAVMs was determined by using CT measurements.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Success and Outcome of Interventions
Complete occlusion with coaxial microcatheters and microcoils was achieved in eight of nine lesions in the first session. Details of the interventional procedures are presented in the Table. The number of coils deployed per lesion (including microcoils and GDCs) varied between 4 and 21 (mean, 10.6 coils). A total of 85 0.018-inch microcoils and 10 GDCs were used in the embolization of the nine PAVMs.
In one patient who had previously undergone an unsuccessful intervention that did not take advantage of the microcatheter technique, a dislodged 0.035-inch (Gianturco) coil impeded vessel catheterization. Only the coaxial microcatheter technique enabled catheterization of the feeder vessel parallel to the misplaced Gianturco coil and allowed for effective microcoil occlusion of the lesion. This patient was the last to be treated with Gianturco coils for PAVM at our institution. From that time on, the microcatheter technique was exclusively used.
One patient who had one PAVM underwent embolization on three different occasions. In this patient, filling of the venous sac alone was performed during the first procedure; the feeding artery was not occluded, which led to incomplete occlusion of the lesion. In the hope that the lesion had spontaneously occluded, a follow-up angiogram was obtained. This angiogram showed residual venous leakage, although the patient had normal coagulatory function. At that time, the distal part of the feeder artery was embolized with microcoils and occlusion was apparently achieved. However, a follow-up angiogram again showed minimal perfusion of the venous component, necessitating a third embolization procedure, during which the total length of the feeding artery was entirely occluded.
Thus, secondary complete occlusion was achieved in all of our patients; none had to undergo surgery for failed embolization with microcoils. No complications as defined in the Materials and Methods section were encountered in any of our patients. In one patient, part of a coil protruded from the occluded feeding artery into a segmental vessel, but no consequences ensued; the vessel remained patent at follow-up angiography. Occasionally, extrasystoles were observed during passage of the guide wire or catheter tip through the right atrium.
Filling of the Venous Sac
In five PAVMs that had a large venous component, the aneurysmal sac was filled with electrolytically detachable GDCs. The large GDCs served as a wire mesh to prevent the embolization of the smaller coils used to occlude the arterial feeding vessel.
Filling of the venous sac alone in one patient (patient 2) did not succeed in occluding the PAVM; therefore, in this patient and in all of the following patients, the feeder artery was embolized with microcoils. In all patients, filling of the venous sac impeded the systemic migration of coils through the PAVM and enabled the subsequent safe occlusion of the feeding artery.
Patient Outcome
In the symptomatic patients who underwent embolization, clinical success was indicated by the complete disappearance of clinical symptoms. Mean supine oxygen saturation significantly (P < .001, t test) increased, from a mean of 90.6% ± 1.8 before embolization to 95.7% ± 2.1 after embolization. All patients were discharged the day of the embolization procedure or the day after the procedure in a state of subjective well-being. All patients underwent follow-up pulmonary angiography 1 day to 7 months after embolization (mean, 2.5 months). Clinical follow-up lasted from 17 to 101 months (mean, 62 months). None of the patients exhibited signs of recanalization of the malformations, dyspnea, or renewed onset of neurologic symptoms.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The use of embolization coils is technically simple; problems such as failure to disengage occur in less than 2% of cases (22). As a rule, several coils are necessary to occlude a PAVM (24), a finding corroborated by our experience. Most authors report excellent long-term results in 90%–95% of cases (4,8,11,16–18,22). An isolated communication (25), however, states that long-term results with (steel) coils were poorer than expected. This finding may be attributable to technical errors. To avoid recanalization, it is important to use coils of appropriate size; a sufficient number of coils must also be inserted tightly into the feeding vessel. If coils are deployed too proximally, collateral perfusion by bronchial artery anastomoses can occur and result in recanalization (11,25).
The literature includes reports of minor complications in 14% of patients and major complications in 4% of patients with PAVM who were treated with embolotherapy (11,20). Minor complications are chest pain and pleurisy, which have a reported frequency of 10%–15% (4,11,20) and are amenable to antiphlogistic drugs. If the embolization technique is adequate, functional loss of lung parenchyma is minimal (20); due to the dual blood supply of the lung, pulmonary infarction occurs only when both pulmonary arteries and bronchial arteries are occluded (16). Systemic migration of embolization materials is one of the most severe complications and occurs in 0.7%–3% of cases (11,21). In one study, stroke caused by migration of embolization materials into the brain vasculature occurred in 1% of treated patients (20). The results of our series compare favorably with those reported in the literature—no complications were observed in any of our patients.
Superselective Embolization
Researchers have hitherto paid little attention to the use of coaxial microcatheters. We began using 3-F coaxial microcatheters in 1992 and have found them useful for superselective placement of the catheter tip directly proximal to the PAVM. The use of microcatheters avoids the risk that a catheter may be dislocated by tension during the advancement of macrocoils or detachable balloons and the subsequent problem of coil deployment in inappropriate vascular territories. The use of microcatheters also eliminates the risk of perforating the venous sac of an aneurysm, a situation that can occur when a 5-F catheter is negotiated into a superselective peripheral position.
Although they did not use such microcoils, Coley and Jackson (22) have argued that 0.018-inch microcoils might be too flexible to be coiled within a PAVM; however, in our experience, they have been easy to handle and highly effective. In our study, the GDCs coiled up properly in the aneurysmal sacs and were not dragged into the systemic circulation, as some authors have speculated could happen (22). The 0.018-inch microcoils also properly engaged with the wall of the feeder arteries in all of our patients. We typically oversized the coils by 10%–20% (usually by 1 mm) to achieve proper engagement.
An important advantage of the use of coaxial microcatheters is that it permits the catheterization of vessels downstream from vessel areas that have already been occluded. This becomes important if coil packing is insufficient and more coils have to be placed. Among our patients, one had a PAVM that was incompletely occluded. Catheterization of the feeding artery with a 5-F catheter was impossible due to a misplaced coil. In this patient, catheterization was possible only with the use of a coaxial microcatheter.
Filling of the Venous Sac
It has been suggested that the risk of systemic coil migration is increased in the presence of a draining vein larger than the feeding artery, a situation in which venous sac occlusion has been proposed (26,27).
According to Takahashi et al (27), filling of the venous sac is beneficial when the draining vessel is larger in diameter than the feeding vessel or when the feeder artery is too short (<15 mm) for secure coiling. For embolization of PAVMs that had a large venous component, we used electrolytically detachable GDCs to prevent systemic coil migration. In our experience, coaxial catheterization facilitates filling of the venous sac. It is still unclear, however, whether filling of the venous sac is actually necessary: At histopathologic analysis, the venous sac may consist of cells divided by a sievelike tangle of thin capillary membranes (9) in which even small coils would become trapped. This applies mainly to complex PAVMs, which we did not treat with filling of the venous sac. Most simple PAVMs appear to consist of a single feeder artery and draining vein with a direct connection (9) through which a small coil might easily pass. Since virtually all PAVMs appear as more or less homogeneous lesions at imaging, it is impossible to determine at imaging whether such a "sieve" is present or not. Therefore, for reasons of safety, we have felt more comfortable with occluding large venous sacs before introducing smaller coils into the feeding artery, especially if a large draining vein is present. Five PAVMs were effectively treated with filling of the venous sac with GDCs.
The primary intention behind venous sac embolization, however, is not to occlude the venous sac itself; rather, it is a preventive measure that facilitates the embolization of the feeding artery. The actual benefit of this remains to be assessed in larger collective studies; however, this study may add to the limited data provided in the literature (26,27).
Use of GDCs
We used GDCs in several cases. An important benefit of electrolytically detachable GDCs, which is shared by other detachable embolization coils such as interlocking detachable coils (IDCs; Boston Scientific/Target Therapeutics) and Jackson detachable coils (Cook), is that they can be retracted if they are deployed in an unfavorable position. The ideal coil size can easily be chosen with detachable coils: A detachable coil used as the first occluding coil can simply be retracted through the insertion catheter if it is the wrong size for a specific vessel (20). It is also possible to reposition the coil if it displays an unfavorable configuration.
Study Limitations and Drawbacks
A drawback of the use of coaxial microcatheters and microcoils is their high cost versus the cost of ordinary (0.035-inch) Gianturco Wallace coils. We deemed their use justified, however, given the rarity of the disease, the potential advantages of the technique, the potentially devastating consequences of systemic coil migration, and the severity of untreated PAVM. A disadvantage of filling the venous sac is the increased number of coils needed to pack the sac compared with the number used in the occlusion of the narrower feeding artery alone (27). We are aware that the value of our findings is limited by the small number of patients treated in our study, a fact that must be attributed to the rarity of the underlying condition.
Recommendations
Patients with PAVM should be treated noninvasively with embolization because up to one-third of patients experience potentially devastating neurologic complications during their lifetime (1,2). Follow-up with chest radiography or CT every 1–2 years has been recommended to exclude the formation of a new PAVM of relevant size (ie, >3–4 mm) (1,2).
Conclusions
In conclusion, the use of coaxial microcatheters and microcoils facilitates the superselective embolization of PAVMs; these devices may also serve as "troubleshooters" in special situations, as the one that occurred in one of our patients, in whom a catheter could only be reinserted into a partially occluded vessel with a coaxial microcatheter. The actual necessity of occluding the venous sac is still unclear; however, it was successful in helping to prevent coil migration in all of our patients.
Future studies that incorporate wide multicenter experience with the treatment of PAVMs with superselective 0.018-inch microcoils and filling of the venous sac are needed to provide definitive data regarding the value of these methods.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, H.P.D., J.T.; study concepts, H.P.D., J.T.; study design, H.P.D.; literature research, H.P.D.; clinical studies, H.P.D., J.T.; data acquisition and analysis/interpretation, H.P.D., J.T.; manuscript preparation, H.P.D.; manuscript definition of intellectual content, H.P.D., J.T.; manuscript editing, H.P.D.; manuscript revision/review and final version approval, H.P.D., J.T.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. I. White Jr, J. S. Pollak, D. Picus, H.-P. Dinkel, and J. Triller Are Guglielmi Detachable Coils Necessary for Treating Pulmonary Arteriovenous Malformations? * Drs Dinkel and Triller respond: Radiology, February 1, 2003; 226(2): 599 - 600. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
