Intracranialdural arteriovenous fistulas (DAVFs) are pathological shunts between dural arteries and dural venous sinuses or leptomeningeal veins. They represent 10%–15% of all intracranial arteriovenous malformations.1Clinical symptomatology varies from asymptomatic lesions to murmur or tinnitus, nonhemorrhagic neurological deficit (NHND), or intracranial hemorrhage and is mainly determined by the venous drainage pattern of the DAVF.2–11The natural history of intracranial DAVFs with cortical venous drainage (CVD) is unfavorable.8,9,12Therefore, DAVFs with CVD should be considered for treatment, as should those without CVD that cause severe clinical symptoms.13–15Endovascular treatment (EVT) is a therapeutic method of choice. Since the introduction of Onyx (ethylene-vinyl alcohol/dimethyl sulfoxide polymer, ev3), transarterial embolization (TAE) of DAVFs using Onyx as the embolic agent has become the preferred treatment strategy.2,16–18然而,缟玛瑙emboliza的长期耐久性tion for DAVFs remains uncertain. The recurrence rate of cured DAVF varies from 0% to 14% after Onyx embolization.19Few studies have reported long-term follow-up, and proposed follow-up programs often differ in choosing between digital subtraction angiography (DSA) and magnetic resonance angiography (MRA).20–22Therefore, we present the most extensive study with long-term follow-up to determine the long-term stability of Onyx and prospectively compare MRA and DSA results to derive an optimal follow-up program for patients after DAVF embolization.
开云体育世界杯赔率
Patient Population
Between January 2002 and December 2018, 156 patients were treated for arteriovenous fistula. All patients underwent diagnostic DSA before definitive diagnosis. Because the study’s focus was on the Onyx embolization technique, 20 patients were excluded because other embolic material, such asN-butyl cyanoacrylate (NBCA) and coils, was used alone. Cases in which Onyx was used with another embolic agent were included in the series. Twenty-four patients with a diagnosis of carotid-cavernous fistula were also excluded.
Clinical Data
Patient demographics were registered, and clinical symptomatology before and after the treatment were compared (Table 1). The number of endovascular procedures, initial angiographic findings, and anatomical location of the DAVF were recorded and analyzed, as well as the embolization route and embolic agent. DAVFs were divided into cured and persistent at the initial and last visits. The length of follow-up, number of controls, and number of DSA and MRA studies performed were recorded (Table 2). Additional treatment for persistent DAVFs was also documented.
Demographics, clinical symptomatology, and outcome
| Value | |
|---|---|
| Demographics | |
| No. of patients | 112 |
| Sex ratio (male/female) | 1/1.73 |
| Mean age, yrs | 60 |
| Presentation | |
| Hemorrhage | 45 (40) |
| NHND | 14 (12.5) |
| Tinnitus or murmur | 25 (22) |
| Headache | 14 (12.5) |
| Vision worsening | 8 (7) |
| Asymptomatic | 9 (8) |
| mRS grade at admission, last follow-up | |
| 0 | 9 (8), 72 (64) |
| 1 | 56 (50), 25 (22) |
| 2 | 20 (18), 3 (2.7) |
| 3 | 12 (11), 5 (4.5) |
| 4 | 7 (6), 4 (3.6) |
| 5 | 8(7), 1(1) |
| 6 | 0 (0), 2 (1.8) |
| Clinical outcome | |
| Improved | 82 (73) |
| No change | 23 (21) |
| Worse | 7 (6) |
Values represent the number of patients (%) unless stated otherwise.
Angiography findings, treatment, and follow-up
| Value | |
|---|---|
| Angiography findings | |
| No. of DAVFs | 113 (100) |
| Borden grade | |
| I | 3 (3) |
| II | 32 (28) |
| III | 78 (69) |
| Anatomical localization | |
| SS/ST | 61 (54) |
| SSS | 18 (16) |
| Tentorium | 12 (11) |
| Anterior fossa | 5 (4) |
| Vein of Galen | 4 (4) |
| Torcula herophili | 3 (3) |
| Other | 10 (9) |
| EVT | |
| No. of procedures | 151 |
| Access route | |
| TAE | 101 (90) |
| TAE/TVE | 11 (10) |
| Embolic agent | |
| Onyx | 89 (79) |
| Onyx/NBCA | 13 (12) |
| Onyx/coils | 8 (7) |
| Onyx/NBCA/coils | 2 (1.8) |
| Outcome | |
| Complete embolization | 98 (87.5) |
| Incomplete embolization | 14 (12.5) |
| Cured DAVF at last follow-up | 105 (94) |
| Follow-up | |
| Mean follow-up | 27.7 |
| Radiological follow-up | |
| MRA only | 54 |
| DSA only | 12 |
| MRA & DSA | 22 |
| Patient followed elsewhere | 5 |
| Lost to follow-up | 24 |
| DAVF recurrence | 2 (1.8) |
SS = sigmoid sinus; SSS = superior sagittal sinus; ST = transverse sinus. Values represent the number of DAVFs (%) unless stated otherwise.
Study Design
The clinical data presented above were retrospectively evaluated. After a retrospective analysis of the series, 15 patients whose DAVF was cured more than 5 years ago were prospectively invited to a clinical control and MRA. It was recommended that these patients undergo confirmative DSA (performed in 10 patients) to evaluate the long-term stability of Onyx and to prospectively compare both diagnostic methods. Our study was conducted in accordance with the Declaration of Helsinki and approved by the institutional ethics committee. All patients gave informed consent.
Endovascular Treatment
All EVTs were performed by one interventional neuroradiologist. The procedures were performed after induction of general anesthesia to minimize patient movement. In all patients, femoral artery access via a 6F sheath (Terumo) was used, and diagnostic angiography starting from an early arterial to late venous phase was performed. Subsequently, through a 6F Chaperon guiding catheter (MicroVention), a superselective catheterization using the microcatheter Marathon 1.5F (ev3), or Sonic 1.2F25 (Balt) with microwire Hybrid 0.007D, 0.008D (Balt), or Mirage (ev3) was performed to reach the nidus of the DAVF. In anatomically challenging situations an intermediate catheter (DAC 0.044, Stryker) to support the stability was used. Superselective angiography was done before every embolization. Onyx-18 (ev3) was our agent of choice in TAE. Onyx was used along with NBCA (B. Braun) in selected cases. Transvenous embolization (TVE) through a common femoral vein with a 6F sheath was used as a secondary option to perform transvenous coil or Onyx application. At the end of the procedure, control angiography was performed.
Results
Patient Population
Of 112 treated patients, 71 (63%) were men and 41 (37%) were women. The mean age at the time of initial EVT was 60 years (range 17–88 years).
Clinical Symptomatology and Outcome
Intracranial hemorrhage (n = 45, 40%), tinnitus or murmur as a dominant symptom (n = 25, 22%), and NHND (n = 14, 12.5%) were the most frequent manifestations of DAVFs. Nine (8%) patients were asymptomatic. Clinical symptomatology was quantified according to the modified Rankin Scale (mRS) grade before EVT and at the last follow-up. Before EVT, 56 (50%) patients had an mRS grade of 1, which was the most common finding. At the last follow-up, there were 72 (64%) asymptomatic patients (i.e., mRS grade 0). Clinical symptomatology and grading before and after the treatment according to mRS grade are shown inTable 1。Two (1.7%) patients died during follow-up due to DAVF-related medical conditions. The first patient underwent two subtotal EVTs of a ruptured DAVF with posterior fossa hemorrhage. One week after a second uncomplicated partial embolization, a progression of hemorrhage occurred, and the patient died as a consequence. The second patient with a Glasgow Coma Scale score of 3 was brought to the hospital after DAVF rupture and died in the intensive care unit despite uncomplicated complete fistula occlusion. At the last follow-up, 82 (73%) patients were clinically improved compared with their initial status, and 23 (21%) remained the same. Periprocedural complications are shown inTable 3。Procedure-related morbidity was reported in 5 (4%) patients. TAE was performed in all patients, who experienced clinical deterioration, one patient underwent both TAE and TVE. All 3 patients with Borden grade I lesions improved clinically. One (3%) and 6 (7.7%) patients with Borden grade II and III DAVFs, respectively, worsened after EVT. Two of 23 (8.7%) patients clinically worsened when more than one embolic agent was used, compared with 5 of 89 (5.6%) who worsened after the use of Onyx alone.
Periprocedural complications
| Patient No. | Age (yrs) | Borden Grade | Location | Ruptured DAVF | Embolization Route | Embolization Material | Complete Embolization | Symptoms | mRS Grade Before EVT | Complication | mRS Grade After EVT |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 17 | III | Tentorium | No | TAE/TVE | Onyx, coils | Yes | Headache | 1 | Hemorrhage | 5 |
| 2 | 61 | II | SS | No | TAE | Onyx | Yes | Tinnitus | 1 | Ischemia | 2 |
| 3 | 30 | III | Tentorium | Yes | TAE | Onyx | Yes | Hemiparesis | 3 | Hemorrhage | 2 |
| 4 | 49 | III | Tentorium | No | TAE | Onyx | Yes | Headache | 1 | Hemorrhage | 0 |
| 5 | 54 | III | SS | Yes | TAE | Onyx | Yes | Headache, diplopia | 1 | Ischemia | 2 |
| 6 | 67 | III | SS | No | TAE | Onyx | Yes | Tinnitus | 1 | Ischemia | 0 |
| 7 | 61 | III | SSS | No | TAE | Onyx | Yes | Paleocerebellar symptomatology | 3 | Ischemia | 4 |
| 8* | 49 | III | SSS | No | TAE | Onyx, NBCA | Yes | Asymptomatic | 0 | Ischemia | 3 |
This patient underwent 2 unsuccessful EVTs elsewhere.
Angiographic Findings, Therapy, and General Follow-Up
Among 112 patients, 113 DAVFs were diagnosed. DAVFs were classified according to the Borden scale. Borden grade III lesions were the most common, being found in 78 (69%) of the patients. The most frequent fistula location was the transverse or sigmoid sinus (61 DAVFs, 54%). The middle meningeal artery was the most common feeder (n = 68, 60%). Borden classification and anatomical localization of DAVFs are given inTable 2。
A total of 151 ETVs were performed for the 113 fistulas (i.e., 1.33 procedures per patient). Representation of access routes and the use of embolic materials are also documented inTable 2。一波登年级我DAVF同时诊断with a Borden grade II DAVF was left without intervention. This fistula spontaneously disappeared without any treatment. In 98 of 112 (87.6%) embolized fistulas, complete occlusion was achieved; 77 (78.5%) of these lesions were embolized in a single EVT session. In 14 (12.5%) cases, partial or subtotal embolization was achieved. TAE was associated with a higher occlusion rate compared with the TAE/TVE combination (88% vs 81%). Incomplete embolization was achieved in 1 (33%) Borden grade I, 4 (12.5%) Borden grade II, and 9 (11.5%) Borden grade III lesions. A higher occlusion rate was recorded when Onyx alone was used (90%) than when multiple embolic agents were used (78%). Four patients underwent Gamma Knife surgery; complete regression of the residual fistula was observed in 3 of these patients. Four patients underwent open surgery: 2 for concomitant pial AVM and 2 for intracerebral hematoma. At the last follow-up, 105 (93%) DAVFs were classified as cured (i.e., completely embolized DAVFs and DAVFs that thrombosed spontaneously or after Gamma Knife surgery). The remaining 8 (7%) DAVFs are being followed after incomplete occlusion. Periprocedural complications were recorded in 8 (7%) patients. An overview of periprocedural complications is given inTable 3。
The overall mean follow-up was 27.7 months (range 0–178 months). The representation of diagnostic methods in the follow-up is listed inTable 2。Progression of 2 residual lesions of previously seemingly cured DAVFs was observed; both were first diagnosed by MRA and confirmed with DSA. One of these lesions underwent final complete embolization (Fig. 1), while the other is being monitored (Fig. 2).
Clinical course of symptomatic DAVF recurrence.A: Borden grade II DAVF of the sigmoid sinus/transverse sinus supplied by the middle and posterior meningeal arteries (arrows) and ascending pharyngeal artery.B: Oblique posterior view of the ascending pharyngeal artery (arrow).C: Initial complete embolization of the DAVF.D: Anteroposterior radiograph.E: Axial MR angiogram showing the DAVF recurrence (arrow) at the sigmoid sinus/jugular vein junction.F: Superselective angiogram showing DAVF recurrence (arrow) supplied by the ascending pharyngeal artery.G: Final EVT of the DAVF. The recurrence is occluded by coils (arrow).H: Anteroposterior radiograph showing the coils (arrow).
Clinical course of an asymptomatic DAVF recurrence within the prospective group.A: Borden grade II DAVF (arrow) supplied by multiple external carotid artery branches.B: Complete embolization of the DAVF.C: MR angiogram showing DAVF recurrence (arrow) seen as contrast-enhancing hyperintense arterial blood in torcula.D: Lateral DS angiogram confirming DAVF recurrence (arrow).
Prospective Group
A prospective group of 15 (of 51) patients with 5 or more years of follow-up after total occlusion of their DAVF underwent MRA and neurological examination. DSA was also performed in 10 of these 15 patients. There was no recurrence of clinical symptoms in the prospective group. However, in 1 patient, MRA showed a residual of a previously seemingly cured DAVF that was later confirmed on DSA without further intervention (Fig. 2). The mean follow-up was 96 months (range 62–178 months) (Table 4). The remaining 36 patients with a sufficiently long follow-up chose not to participate, could not be reached, or died of DAVF-unrelated causes.
Prospective group
| Patient No. | Sex | Presenting Complaints | mRS Grade (admission/last FU) | Borden Grade | Location | No. of Procedures | Access Route | Embolic Agent | Complete Embolization | Periprocedural Complications | MRA Performed | DSA Performed | FU (mos) | DAVF Recurrence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | F | Vision worsening | 2/0 | II | CS | 2 | TAE | Onyx | Yes | No | Yes | No | 112 | No |
| 2 | F | Tinnitus | 2/0 | II | Torcula | 4 | TAE/TVE | Onyx | Yes | No | Yes | No | 92 | No |
| 3 | F | NHND | 2/0 | III | SSS | 1 | TAE | Onyx | Yes | No | Yes | No | 72 | No |
| 4 | M | Hemorrhage | 1/0 | III | SS/ST | 1 | TAE | Onyx | Yes | No | Yes | No | 64 | No |
| 5 | F | Asympt | 0/0 | III | Tentorium | 1 | TAE | Onyx | Yes | No | Yes | Yes | 104 | No |
| 6 | M | Tinnitus | 1/0 | II | SS/ST | 1 | TAE | Onyx | Yes | No | Yes | Yes | 97 | Yes |
| 7 | M | NHND | 2/0 | II | SS/ST | 1 | TAE | Onyx | Yes | No | Yes | Yes | 94 | No |
| 8 | M | Asympt | 0/0 | II | SS/ST | 1 | TAE | Onyx | Yes | No | Yes | Yes | 102 | No |
| 9 | M | Hemorrhage | 1/0 | III | SS/ST | 1 | TAE | Onyx | Yes | No | Yes | Yes | 92 | No |
| 10 | M | NHND | 1/0 | III | Anterior fossa | 1 | TAE | Onyx | Yes | No | Yes | Yes | 98 | No |
| 11 | M | Asympt | 0/0 | III | SSS | 1 | TAE | Onyx | Yes | No | Yes | Yes | 65 | No |
| 12 | M | NHND | 2/0 | III | SS/ST | 3 | TAE | Onyx, NBCA | Yes | No | Yes | Yes | 178 | No |
| 13 | M | Tinnitus | 1/0 | III | SS/ST | 3 | TAE/TVE | Onyx, coils | Yes | No | Yes | Yes | 146 | No |
| 14 | M | Headache | 1/1 | III | SSS | 1 | TAE | Onyx | Yes | No | Yes | No | 62 | No |
| 15 | M | Headache | 1/1 | II | Tentorium | 1 | TAE | Onyx | Yes | No | Yes | Yes | 67 | No |
Asympt = asymptomatic; CS = cavernous sinus; FU = follow-up.
Discussion
DAVFs are the most common acquired intracranial vascular lesions.23They may be completely asymptomatic, or they may cause fatal intracranial bleeding. The key risk factor for hemorrhage or NHND is the presence of CVD or venous varix.3,6,7,24,25Several studies have described the natural course of DAVF with CVD.3,8,9,12,26Gross and Du, for instance, reported a 6% annual hemorrhage risk for DAVFs with indirect CVD and a 10% annual hemorrhage risk for DAVFs with direct CVD, increasing to 21% if a venous varix is present.12In a series of 20 patients with DAVF rupture, Duffau et al. reported 3 deaths, 1 impairment, and 16 (80%) cases of a favorable clinical course.9Asymptomatic lesions are considered much less aggressive than those initially presenting with bleeding or NHND.12,26Previous topographical classifications3,27–29have been replaced by classifications based on angiographic findings that demonstrate the risk of eventual hemorrhage or NHND.6,7,24,25,30,31The Borden and Cognard grading systems are the most frequently used scales.6,7High-risk DAVFs are indicated for treatment, as well as low-risk fistulas causing significant clinical symptoms. There are various treatment methods, including EVT, open surgery, and radiosurgery. Observation is also a relevant modality for asymptomatic DAVF without CVD; however, patients should be regularly monitored for the risk of developing CVD.4Gross et al. reported no bleeding or NHND in 126 Borden grade I DAVFs over 177 lesion-years.32Among the mentioned methods, EVT is currently the first-choice modality. TAE with Onyx as an embolic agent is the preferred method of DAVF treatment at most institutions.2,16–18The therapeutic goals are to close the shunt at the site of the fistula and to extend the embolic material into the venous side.21
Over 16 years, 112 patients underwent 151 EVTs for 113 DAVFs in our department. After incomplete embolization, 4 patients were indicated for Gamma Knife surgery. Only 4 patients in the series underwent open surgery. Although EVT is considered the method of choice, open surgical treatment of DAVFs remains a versatile and effective option for the management of various aggressive intracranial DAVFs.33While our approach to surgical treatment of pial AVMs is aggressive, the representation of microsurgical intervention in DAVF therapy is relatively low in our cohort. Moenninghoff et al. reported surgical treatment of DAVFs in 9%, Oh et al. in 13%, and Ambekar et al. in 38% of cases.13,34,35Our rate of microsurgical intervention is low because we have good long-term experience with EVT using Onyx, a low recurrence rate, and minimal procedure-related complications. Furthermore, because we are considered the regional center for EVT of DAVFs, high-risk inoperable lesions are concentrated in our clinic.
A high percentage of DAVFs with CVD indicates that neurological deficit induced by intracranial bleeding was the most common cause of hospitalization (40%) at our institution. Tinnitus and murmur (22%) and NHND (12.5%) were less common. Although in other studies DAVFs with direct or indirect leptomeningeal vein drainage were frequent, hemorrhage was observed less often than in our study.13,18,34,36Moenninghoff et al. diagnosed CVD in 71% of their cases but bleeding in only 13%.34In their series, Hu et al. reported that 57% of their patients were diagnosed with CVD and 22% with hemorrhage.36In our study, clinical symptomatology was classified according to the mRS grading system (Table 1). Overall, 73% of our patients clinically improved after EVT and 21% went unchanged. Negative prognostic factors in our series were the TAE access route and the higher initial Borden grade. The prognostic difference between the sole use of Onyx and multiagent embolization was insignificant (5.6% vs 8.7%). However, our data may be biased due to the small sample in which the TAE/TVE (10%) access route was used and relatively low representation of adjunctive embolysates (28%). In their series, Oh et al. did not notice a significant prognostic difference between TAE and TAE/TVE.13The same study also found no difference between the use of Onyx and Onyx with adjunctive embolysates. However, a higher Borden grade as a negative prognostic factor is well documented in the literature. Gross et al., for instance, documented 6% of permanent neurological complications in high-grade DAVFs compared with 3% in the entire series.17
Borden classification, anatomical localization, and embolization route as well as the embolic agent and angiographic outcome are given inTable 2。Borden grade III was the most common type of DAVF, with an incidence rate of 69%. Other series have reported a lower incidence of Borden grade III DAVFs (Chandra et al., 63%;22Natarajan et al., 41%;2and Oh et al., 26%13). As expected, anatomical localization was in agreement with the literature, where the transverse sigmoid sinus location has been reported in 50% of cases, tentorium in 12%, and superior sagittal sinus in 8%.3,6,7,37By embolization, we were able to cure 87.5% of the DAVFs. The literature shows considerable heterogeneity in the initial occlusion rate, ranging from 47% to 100%.19In a systematic review and meta-analysis of studies describing TAE using Onyx, Sadeh-Gonike et al. reported an overall initial complete occlusion rate of 82%.19Periprocedural complications were recorded in 8 (7%) patients (5 cases of ischemia and 3 of hemorrhage). However, 3 of those complications were clinically silent (Table 3). Sadeh-Gonike et al. reported a pooled procedure-related morbidity rate of 3%.19Our results confirm that, when using Onyx, TAE is a very effective and safe treatment modality for DAVF. In indicated cases, the use of other materials (in particular, NBCA and coils), as well as TVE, may be beneficial in achieving complete occlusion. However, the long-term stability of Onyx has been repeatedly questioned in the past,16,21,36,38,39and recent studies with recurrence rates ranging from 0% to 14.3% do not provide a definitive answer.19
红玛瑙是一种液体栓塞共聚物组成的ethylene-vinyl alcohol as an embolic agent dissolved in dimethyl sulfoxide and tantalum powder for radiological visualization. In 2007, Carlson et al. were the first to publish a successful DAVF occlusion using Onyx.40Ethylene-vinyl alcohol has clear advantages over other embolic agents, especially NBCA. Its nonadhesive behavior allows a slow, controlled application. Slower precipitation time makes the penetration of the agent into the venous part of the fistula more likely, and the possibility of applying a larger amount of embolic agent allows for reflux into other arterial feeders in single-feeder embolization. Since its introduction, there have been several studies showing the superiority of Onyx over NBCA.17,41,42In a series of 56 DAVFs, Rabinov et al. presented a higher occlusion rate of 83% in Onyx-treated lesions compared with 33% when NBCA was used.41In our study, we also observed a lower occlusion rate when using more embolic materials compared with the sole use of Onyx (78% vs 90%). The same conclusion was reached by Kim et al., who reported incomplete occlusion in 37.5% of cases in which adjunctive embolysates were used compared with 19% when only Onyx was used.43However, long-term occlusion depends on multiple factors and not only the embolic agent. It is necessary to extend the embolic material into the venous side of DAVF.21Another important factor is the access route. Oh et al. presented complete or near-complete occlusion in 42% of DAVFs treated with TAE and in 84% of DAVFs treated with TVE.13At our clinic, TAE is the preferred embolization method. Based on long-term experience, we were able to achieve complete embolization in 88% of patients solely with TAE. In cases in which the combined TAE/TVE approach was used, the percentage of complete embolization was lower (81%). The angioarchitecture of a DAVF itself is another important predictive factor for successful occlusion. Vollherbst et al. stated that the presence of multiple feeding arteries and involvement of the pharyngeal artery negatively influence the treatment success.44We did not notice a correlation between Borden grade and occlusion rate in our study, as the low occlusion rate in Borden grade I DAVFs (33%) is probably biased due to the small number of these lesions.
There have been some concerns about Onyx use. First, Onyx may cause vascular inflammation leading to cranial nerve palsies (primarily in carotid-cavernous fistulas).16Second, there is a question about its long-term durability. In our group of 98 DAVFs in which we were able to achieve complete embolization, we experienced two recurrences. In both cases, the Onyx material remained stable; however, the embolization caused hemodynamic changes that led to the accentuation of residual arterial feeders not visible on immediate postembolization DSA. One of 2 patients developed tinnitus that resulted in diagnostic MRA and subsequent DSA with final embolization in 1 session. The other patient with recurrence was completely asymptomatic for 8 years. The residual fistula was diagnosed in the context of the prospective group.
The prospective group of patients with complete fistula occlusion more than 5 years earlier was first established to determine the long-term stability of Onyx, and second, to compare MRA and DSA as the follow-up diagnostic methods (Table 4). Of a group of 15 patients, all underwent clinical examination and MRA. DSA was performed in 10 of these patients; the remaining 5 refused to undergo DSA. In the entire prospective group, there was no relapse of clinical symptoms. However, in 1 patient there was a slight residual progression of the occluded DAVF of the SS/ST junction diagnosed on MRA and later confirmed on DSA without further intervention. The analysis of the postembolization DSA indicated that the residual arterial feeder was not initially apparent. At the control DSA within the prospective group, this feeder was clearly visible with residual fistula shunting to the torcular Herophili. This leads to the hypothesis that the elimination of major arterial feeders can result in hemodynamic changes that accentuate residual feeders not noticeable on immediate postembolization images. Because of the insignificant residual size and the absence of clinical symptoms, the patient was not indicated for reintervention and is still being monitored.
Although the mechanism of recurrence of DAVFs after Onyx embolization remains unclear, the literature offers several possible explanations. The first describes that Onyx can leave microchannels within the cast that allow small amounts of flow through the fistula, resulting in recurrence of DAVF.16The second is based on the lack of penetration of Onyx into the fistula, which leads to recruitment or enlargement of the previously not shown feeder.20,45The third suggests that embolization leads to the development of hemodynamic changes that results in the appearance of the part of the fistula that was previously masked by a high-flow shunt.20,36Based on our results, we believe in the long-term stability of Onyx as embolic material. In both cases of DAVF recurrence in our series, a similar mechanism of development based on the described change in hemodynamic conditions occurred (Figs. 1and2). Onyx, as the embolic material, remained stable in both cases; however, progressions of vascular connections below the resolution of the initial angiogram were observed. Therefore, from our point of view, a more accurate designation than “recurrence” would be “progression of a primarily graphically masked residue.”
We observed DAVF progression of a primarily graphically masked residual in 1.8% of the patients in the entire series. Considering the long-term follow-up and that this is one of the largest clinical series to date, we believe that the risk of recurrence is exceptionally low. Although the literature presents a 2% risk of recurrence after TAE with Onyx, this figure refers to midterm follow-up (mean 5 months) studies.19There are few studies with long-term follow-up.20–22而钱德勒等人短期recurre报道nce rate of 6% but long-term stability of Onyx embolization, Ambekar et al. reported recurrence of DAVF in 14.3% of their patients at a mean follow-up of 14 months.21,22Rangel-Castilla et al. reported that the stability of obliteration at 6, 12, 24, and 46 months was 100%, 95.4%, 93.8%, and 92.3%, respectively, after total or near-total obliteration.20这些研究得出的结论是来讲adictory and inconsistent. Ambekar et al.21highly recommend long-term angiographic follow-up. However, Chandra et al.22suggested performing DSA in the first 6 months, which, if complete occlusion is confirmed, would be followed by repeat MRA controls. To the best of our knowledge, we present the first prospective series of DAVFs that compares MRA and DSA as the follow-up diagnostic methods. Both of our DAVF recurrences were first diagnosed on MRA and confirmed on DSA. Based on these results, we believe that MRA could be a sufficient diagnostic modality for follow-up of patients after complete DAVF embolization. This leads us to propose a modified follow-up program.
After initial complete graphic occlusion of DAVF, MRA and clinical examination will be performed during further follow-up controls: 6 months after the procedure and then at 12, 24, 60, and 120 months after complete occlusion of the DAVF. In case of relapse of clinical symptoms or unclear findings on MRA, DSA will be performed as soon as possible. The advantage of the modified program is the elimination of risks associated with angiography, either local (inguinal hematoma, pseudoaneurysm of the femoral artery) or, above all, central ischemic stroke. Ultimately, we eliminate the need for short-term hospitalization when the patient undergoes DSA.
In cases of incomplete embolization of the DAVF, we would recommend DSA 6 months after the initial procedure, then every 12 months for 3 years and, if no progression is observed, every 24 months. Early DSA should be performed on relapse/progression of clinical symptoms. Open surgery or radiosurgery can be considered as a secondary option to achieve complete fistula occlusion. Once DAVF occlusion after open surgery or radiosurgery is apparent on DSA, the patient will undergo clinical follow-up and MRA, as mentioned above. Long-term follow-up should be considered in all patients with complex DAVFs.
The results of the present series show that MRA is a sufficient follow-up method after complete DAVF occlusion. However, we acknowledge that our study has its limits, mentioned below. Although we believe that the proposed follow-up is reasonable, we are aware that at this point it is not possible to implement it in clinical practice. Its validity must be confirmed by prospective studies with standardized follow-up, in which all patients will undergo MRA and DSA at defined time intervals. We hope that the presented prospective substudy is an impulse for such future series.
Strengths and Weaknesses
We present the largest series of DAVFs with long-term follow-up and the first study comparing MRA and DSA as follow-up diagnostic methods. Such an approach allows demonstration of the long-term stability of Onyx. Most of the population was evaluated retrospectively. However, the prospective subgroup gives our findings greater scientific validity.
One of the weaknesses of the study is the high variability in follow-up, which ranged from 0 to 178 months. Also, a larger percentage (21%) of our patients were lost to follow-up and only 30% of our series underwent DSA after DAVF embolization.
The main weakness of our series is the small size of the prospective group based on which we are unable to definitively determine the best follow-up diagnostic method. Further larger prospective studies must be conducted.
Conclusions
我们的研究证实,TAE缟玛瑙是一个高度effective and safe therapeutic method for the treatment of DAVFs. We believe that Onyx is a long-term stable embolic material. However, the recurrences of seemingly completely embolized DAVFs may occur (although rarely) because of postembolization hemodynamic changes that accentuate residual fistulas. Based on our results, we believe that these residuals are evident on MRA. Further prospective studies should confirm whether MRA is a sufficient diagnostic method for the follow-up of DAVF after total occlusion.
Acknowledgments
Our study was supported by the following grants: Q28/LF1 and MO IP 1012 from Charles University.
Disclosures
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
Author Contributions
Conception and design: all authors. Acquisition of data: Voldřich, Charvát. Analysis and interpretation of data: all authors. Drafting the article: Voldřich, Netuka, Beneš. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Voldřich. Administrative/technical/material support: Voldřich. Study supervision: Netuka, Charvát, Beneš.
References
-
1 ↑
NewtonTH,CronqvistS。Involvement of dural arteries in intracranial arteriovenous malformations。Radiology。1969;93(5):1071–1078。
-
2 ↑
NatarajanSK,GhodkeB,KimLJ,et al.Multimodality treatment of intracranial dural arteriovenous fistulas in the Onyx era: a single center experience。World Neurosurg。2010;73(4):365–379。
-
3 ↑
AwadIA,LittleJR,AkarawiWP,AhlJ。Intracranial dural arteriovenous malformations: factors predisposing to an aggressive neurological course。J Neurosurg。1990;72(6):839–850。
-
4 ↑
CognardC,HoudartE,CasascoA,et al.Long-term changes in intracranial dural arteriovenous fistulae leading to worsening in the type of venous drainage。Neuroradiology。1997;39(1):59–66。
-
5
AndrewsBT,DujovnyM,MirchandaniHG,AusmanJI。Microsurgical anatomy of the venous drainage into the superior sagittal sinus。开云体育app官方网站下载入口。1989;24(4):514–520。
-
6 ↑
BordenJA,WuJK,ShucartWA。A proposed classification for spinal and cranial dural arteriovenous fistulous malformations and implications for treatment。J Neurosurg。1995;82(2):166–179。
-
7 ↑
CognardC,GobinYP,PierotL,et al.Cerebral dural arteriovenous fistulas: clinical and angiographic correlation with a revised classification of venous drainage。Radiology。1995;194(3):671–680。
-
8 ↑
DaviesMA,Ter BruggeK,WillinskyR,WallaceMC。The natural history and management of intracranial dural arteriovenous fistulae. Part 2: aggressive lesions。Interv Neuroradiol。1997;3(4):303–311。
-
9 ↑
DuffauH,LopesM,JanosevicV,et al.Early rebleeding from intracranial dural arteriovenous fistulas: report of 20 cases and review of the literature。J Neurosurg。1999;90(1):78–84。
-
10
GistTL,Rangel-CastillaL,KrishnaC,et al.Endovascular management of six simultaneous intracranial dural arteriovenous fistulas in a single patient。J Neurointerv Surg。2014;6(2):e16。
-
11
OsbunJW,KimLJ,SpetzlerRF,McDougallCG。Aberrant venous drainage pattern in a medial sphenoid wing dural arteriovenous fistula: a case report and review of the literature。World Neurosurg。2013;80(6):e381–e386。
-
12 ↑
GrossBA,DuR。The natural history of cerebral dural arteriovenous fistulae。开云体育app官方网站下载入口。2012;71(3):594–603。
-
13 ↑
OhSH,ChoiJH,KimBS,et al.Treatment outcomes according to various treatment modalities for intracranial dural arteriovenous fistulas in the Onyx era: a 10-year single-center experience。World Neurosurg。2019;126:e825–e834。
-
14
ChoiJH,JoKI,KimKH,et al.Spontaneous angiographic changes in venous drainage patterns related to symptom changes in patients with untreated cavernous sinus dural arteriovenous fistula。Neuroradiology。2015;57(11):1153–1161。
-
15
GemmeteJJ,AnsariSA,GandhiD。Endovascular treatment of carotid cavernous fistulas。Neuroimaging Clin N Am。2009;19(2):241–255。
-
16 ↑
GhobrialGM,MarchanE,NairAK,et al.硬脑膜动静脉瘘管:lite的审查rature and a presentation of a single institution’s experience。World Neurosurg。2013;80(1-2):94–102。
-
17 ↑
GrossBA,AlbuquerqueFC,MoonK,McDougallCG。Evolution of treatment and a detailed analysis of occlusion, recurrence, and clinical outcomes in an endovascular library of 260 dural arteriovenous fistulas。J Neurosurg。2017;126(6):1884–1893。
-
18 ↑
JohnsonCS,ChiuA,CheungA,WenderothJ。Embolization of cranial dural arteriovenous fistulas in the liquid embolic era: a Sydney experience。J Clin Neurosci。2018;49:62–70。
-
19 ↑
Sadeh-GonikeU,MagandN,ArmoiryX,et al.Transarterial Onyx embolization of intracranial dural fistulas: a prospective cohort, systematic review, and meta-analysis。开云体育app官方网站下载入口。2018;82(6):854–863。
-
20 ↑
Rangel-CastillaL,BarberSM,KlucznikR,DiazO。Mid and long term outcomes of dural arteriovenous fistula endovascular management with Onyx. Experience of a single tertiary center。J Neurointerv Surg。2014;6(8):607–613。
-
21 ↑
AmbekarS,GaynorBG,PetersonEC,ElhammadyMS。Long-term angiographic results of endovascularly “cured” intracranial dural arteriovenous fistulas。J Neurosurg。2016;124(4):1123–1127。
-
22 ↑
ChandraRV,Leslie-MazwiTM,MehtaBP,et al.Transarterial Onyx embolization of cranial dural arteriovenous fistulas: long-term follow-up。AJNR Am J Neuroradiol。2014;35(9):1793–1797。
-
23 ↑
WankeI,RüfenachtDA。The dural AV-fistula (DAVF), the most frequent acquired vascular malformation of the central nervous system (CNS)。Clin Neuroradiol。2015;25(suppl 2):325–332。
-
24 ↑
ZipfelGJ,ShahMN,RefaiD,et al.Cranial dural arteriovenous fistulas: modification of angiographic classification scales based on new natural history data。Neurosurg Focus。2009;26(5):E14。
-
25 ↑
BaltsaviasG,RothP,ValavanisA。Cranial dural arteriovenous shunts. Part 3. Classification based on the leptomeningeal venous drainage。Neurosurg Rev。2015;38(2):273–281。
-
26 ↑
DaviesMA,SalehJ,Ter BruggeK,et al.The natural history and management of intracranial dural arteriovenous fistulae. Part 1: benign lesions。Interv Neuroradiol。1997;3(4):295–302。
-
27
HayesGJ。Externa carotid-cavernous sinus fistulas。J Neurosurg。1963;20:692–700。
-
28
AminoffMJ。血管异常the intracranial dura mater。Brain。1973;96(3):601–612。
-
29
PicardL,BracardS,MoretJ,et al.Spontaneous dural arteriovenous fistulas。Semin Intervent Radiol。1987;4:219–241。
-
30 ↑
CastaigneP,BoriesJ,BrunetP,et al.Meningeal arterio-venous fistulas with cortical venous drainage. Article in French。Rev Neurol (Paris)。1976;132(3):169–181。
-
31 ↑
DjindjianR,MerlandJJ。Normal super-selective arteriography of the external carotid artery。In:Super-Selective Arteriography of the External Carotid Artery。Springer;1978:1–123。
-
32 ↑
GrossBA,AlbuquerqueFC,McDougallCG,et al.A multi-institutional analysis of the untreated course of cerebral dural arteriovenous fistulas。J Neurosurg。2018;129(5):1114–1119。
-
33 ↑
ColliceM,D’AlibertiG,ArenaO,et al.Surgical treatment of intracranial dural arteriovenous fistulae: role of venous drainage。开云体育app官方网站下载入口。2000;47(1):56–67。
-
34 ↑
MoenninghoffC,PohlE,DeuschlC,et al.Outcomes after Onyx embolization as primary treatment for cranial dural arteriovenous fistula in the past decade。Acad Radiol。2020;27(6):e123–e131。
-
35 ↑
AmbekarS,GaynorBG,PetersonEC,ElhammadyMS。Long-term angiographic results of endovascularly “cured” intracranial dural arteriovenous fistulas。J Neurosurg。2016;124(4):1123–1127。
-
36 ↑
HuYC,NewmanCB,DashtiSR,et al.Cranial dural arteriovenous fistula: transarterial Onyx embolization experience and technical nuances。J Neurointerv Surg。2011;3(1):5–13。
-
37 ↑
KiyosueH,HoriY,OkaharaM,et al.Treatment of intracranial dural arteriovenous fistulas: current strategies based on location and hemodynamics, and alternative techniques of transcatheter embolization。射线照相。2004;24(6):1637–1653。
-
38 ↑
AdamczykP,AmarAP,MackWJ,LarsenDW。Recurrence of “cured” dural arteriovenous fistulas after Onyx embolization。Neurosurg Focus。2012;32(5):E12。
-
39 ↑
LuoCB,ChangFC,Mu-HuoTeng M,et al.Transarterial Onyx embolization of intracranial dural arteriovenous fistulas: a single center experience。J Chin Med Assoc。2014;77(4):184–189。
-
40 ↑
CarlsonAP,TaylorCL,YonasH。Treatment of dural arteriovenous fistula using ethylene vinyl alcohol (Onyx) arterial embolization as the primary modality: short-term results。J Neurosurg。2007;107(6):1120–1125。
-
41 ↑
RabinovJD,YooAJ,OgilvyCS,et al.ONYX versus n-BCA for embolization of cranial dural arteriovenous fistulas。J Neurointerv Surg。2013;5(4):306–310。
-
42 ↑
ChooDMSJ,ShankarJJ。Onyx versus nBCA and coils in the treatment of intracranial dural arteriovenous fistulas。Interv Neuroradiol。2016;22(2):212–216。
-
43 ↑
KimB,JeonP,KimK,et al.Predictive factors for response of intracranial dural arteriovenous fistulas to transarterial Onyx embolization: angiographic subgroup analysis of treatment outcomes。World Neurosurg。2016;88:609–618。
-
44 ↑
VollherbstDF,HerwehC,SchönenbergerS,et al.The influence of angioarchitectural features on the success of endovascular embolization of cranial dural arteriovenous fistulas with Onyx。AJNR Am J Neuroradiol。2019;40(12):2130–2136。
-
45 ↑
NogueiraRG,DabusG,RabinovJD,et al.Preliminary experience with onyx embolization for the treatment of intracranial dural arteriovenous fistulas。AJNR Am J Neuroradiol。2008;29(1):91–97。
