Single-center Retrospective Analysis of One-stop Hybrid Surgery for Brain Arteriovenous Malformations

Objectives To evaluate the safety and effectiveness of one-stop hybrid surgery of microsurgical resection combined with preoperative embolization or intraoperative angiography for brain arteriovenous malformations (bAVMs) and compare the ecacy of one-stop hybrid surgery with nonhybrid surgery (endovascular embolization alone) in the treatment of bAVMs. Methods A single-center retrospective study was performed to recruit 70 patients with bAVMs at our hospital between July 2017 and April 2020. Patients were divided into 2 groups depending on the mode of surgery: hybrid group (i.e., microsurgical resection combined with preoperative endovascular embolization or intraoperative angiography in the hybrid operating room), and nonhybrid group (i.e., endovascular embolization, stereotactic radiosurgery, or microsurgical resection alone). The hybrid group was divided into two subgroups: the microsurgical resection combined with preoperative embolization, and the microsurgical resection combined with intraoperative angiography subgroups. All patients’ demographic variables, clinical manifestations, and imaging features, postoperative complications, and long-term clinical prognosis were recorded and analyzed. or nonhybrid treat bAVMs clinically, whereas further follow-up is needed to determine the long-term effects after operation.


Introduction
Brain arteriovenous malformations (bAVMs) refer to the congenital malformation with direct arteriovenous communications characterized by an abnormal connection between feeding arteries and draining veins lack of the brain capillary network [17]. However, the accurate etiology and pathogenesis of bAVMs remain inconsistent. The potential mechanism may be attributed to congenital aberrancies of vascular development in the embryo. bAVMs usually occur in young adults between the age of 20 and 40 years old, with an average age of 34 years [11]. The most commonly observed neurological presentation is intracranial hemorrhage (3%), followed by focal or generalized seizure (1%), chronic headache, progressive neurological de cit, or no symptoms [3,4,6,7]. To achieve the goal of preventing cerebral hemorrhage and decreasing the frequency of seizures, various treatments have emerged in an endless stream in recent years.
The major treatment modalities for bAVMs including endovascular embolization, microsurgical resection, stereotactic radiosurgery, either alone or in combination to improve the rate of complete nidal obliteration [37]. Generally, microsurgical resection is recommended as a treatment modality for patients with low-grade bAVMs (grade I-II) [24]. Moreover, endovascular embolization is now recognized as an effective treatment performed alone or combined with microsurgical resection [14,35]. However, three traditional treatments for bAVMs have the limitations that approximately 30% of patients remain untreatable despite successfully eliminating the lesion.
In the past ten years, further studies on the treatment of bAVMs tend to be multimodality treatment [10,22,23]. A recent expert consensus on the treatment of bAVMs suggested that hybrid intervention of microsurgery, intraoperative angiography, and endovascular embolization in a hybrid operating room may be an advanced and effective approach for the management of bAVMs [13]. Although a few studies have con rmed its effectiveness and bene ts such as a high rate of immediate, complete, and permanent AVM obliteration, its exact e cacy has yet to be completely revealed [9,21,33]. Here, we report a series of bAVMs treated with a hybrid intervention or endovascular treatment alone. Furthermore, we compared the clinical outcome and prognosis between hybrid operation and nonhybrid operation (i.e. endovascular embolization). and introduced the therapeutic concepts of hybrid surgery for bAVMs.

Patients
In the present study, we conducted a retrospective analysis of the continuous clinical data of 70 patients with cerebral AVMs who underwent surgical treatment in the Department of Neurosurgery, Zhongnan Hospital of Wuhan University from July 2017 to April 2020. To reduce selection bias, we excluded patients who did not receive any history of AVM treatment during the study period. All Patients were classi ed into two main groups: hybrid surgery group (pre-operative embolization/intraoperative digital subtraction angiography (DSA) with subsequent microsurgical resection) and nonhybrid surgery group (endovascular embolization alone). Thus, we compared the baseline clinical data and clinical outcomes between the hybrid operation group and the endovascular embolization alone group.

Clinical data and Radiological features
The general data analyzed mainly included gender, age, rupture status, clinical symptoms (i.e. headache, seizure, neurological de cits, incidental ndings), and all patients underwent DSA and MRI for initial diagnosis. The characteristics of bAVM including Spetzler-Martin (SM) score, location of the bAVM, functional area, the pattern of venous drainage for each patient were assessed using MRI and DSA. The AVM lesions were evaluated based on morphologic characteristics, according to the SM grading system [30].
Treatment bAVM resection and preoperative embolization were completed at a single stage in all. Both endovascular embolization and intravascular embolization was performed by using a biplane at-panel angiographic suite (UNIQ FD2020 Hybrid-OR, Philips, Eindhoven, The Netherlands) with 3D reconstruction under general anesthesia. Intraoperative DSA is generally performed during the operation and before craniotomy closure to con rm whether the bAVM lesion is completely removed. Presurgical embolization was performed by using either n-butyl cyanoacrylate (Tru ll, Cordis Neurovascular, Inc., Miami Lakes, FL, USA) or ethylene-vinyl copolymer (Onyx 18 or 34, Medtronic, Inc., Minneapolis, Minnesota, USA), detachable coils. The whole operation was carried out in a sterile environment with the use of sterile techniques, without changing the position of the patient during hybrid surgery or endovascular embolization alone.

Outcome assessment
Prognostic factors of functional outcome were assessed based on the modi ed Rankin Scale (mRS). The mRS score was estimated immediately before and one week after the treatment of intravascular embolization or hybrid operation. The neurological functional outcome was dichotomized into 2 groups: mRS score ≤ 2 was regarded as a good prognosis, whereas mRS score > 2 was de ned as a poor prognosis. The change value of mRS was evaluated in postoperative vs. preoperative and follow-up vs. preoperative. Clinical follow-up for all patients averaged 6 months. Long-term follow-up was through regular clinical follow-up postoperative clinical visits, hospitalization, or a combination of telephone interviews obtained.

Statistical analysis
All normally distributed continuous variables were reported as mean ± standard deviation (SD), and categorical data as percentages or frequencies for descriptive characteristics. We assessed the differences between the two groups by t-tests or Mann-Whitney U for continuous variables and χ 2 or Fisher exact tests for continuous and categorical variables, respectively. The statistical signi cance level was set at P ≤ 0.05, and all data were analyzed using IBM SPSS Statistics for Windows, version 25.0 (IBM Corp., Armonk, New York, USA) for statistical analysis.

Ethics approval
All procedures performed in studies involving human participants were in accordance with the 2013 Helsinki declaration. This retrospective study was approved by the Ethics Committee of Zhongnan Hospital of Wuhan University. At the same time, we have already obtained informed consent from all included patients. This article does not contain any studies with human participants or animals performed by any of the authors.
By subgroup comparisons in hybrid group, there were no signi cant differences in terms of age (P = 0.438), gender (P = 0.635), and preoperative mRS (P = 0.109), except for ruptured AVMs (P = 0.013). The detailed information is presented in Table 2.

AVM Characteristics
There were 7 cases in SM grade I, 25 grade II, 26 grade III, 11 grade IV, and 1 grade V AVMs, according to Spetzler-Martin score ( Figure 2). Nidus sites were localized on supratentorial brain areas in 34 patients (94.44%) from the hybrid group and in 29 cases (85.29%) from the nonhybrid group. The functional areas of the brain were involved in 14 cases (38.89%) from the hybrid group and in 18 cases (52.94%) from the nonhybrid group. However, there was no statistically signi cant difference in these results (Table 1). In the hybrid subgroup, there were no differences between the two groups regarding SM score, nidus site, functional area, and venous drainage ( Table 2).

Treatment and clinical outcomes
Immediate post-procedural angiography indicated angiographic cure rate in 94.44% of patients from the hybrid group, whereas 35.29% were from the nonhybrid group (P<0.001, Table 3). Intraoperative DSA shows that residual AVM nidus was completely excised during this same period in the hybrid group. The majority (66/70) patients (94.3%) had short term follow-up (>90 days) in these series (Tables 3 and 4). The rates of a good outcome were 94.1% in the hybrid group and 90.6% in the nonhybrid group, although there was no statistically signi cant difference between patients with good or unfavorable outcomes (Table 3).
Operation-related general complications occurred in 13 patients (36.1%) in the hybrid group and 5 patients (14.7%) in the nonhybrid group (P = 0.041, Table 3). Among the hybrid group, the rate of procedure-related complications in the presurgical embolization (Embolization + Surgery) subgroup is signi cantly higher compared with the radiography and surgery (Radiography + Surgery) subgroup (P = 0.008, Table 4). Preoperative embolization worsened the clinical condition of 7 patients (19.4%) through a delayed postoperative intracerebral hemorrhage, while 4 patients were received emergency hematoma evacuation. Among the patients with embolization alone, embolization worsened the clinical condition in 3 patients (8.8%) through a delayed post-embolization intracerebral hemorrhage (Table 2). One 39-yearold patient in the embolization alone group died from rehemorrhage after one-year of treatment. No patients died in the hybrid operation group. Among the patients in hybrid group, presurgical embolization was performed in 55.6% of patients and achieved a good clinical outcome (Figure 3, 4).

Discussion
Cerebral AVMs are important causes of intracerebral hemorrhage in young adults. For previously unruptured brain AVMs, the rate of developing hemorrhage was approximately 1% per year, but increases by ve folds once ruptured [15,19,20]. Our single-center retrospective study showed 48 patients (68.6%) presenting ruptured AVMs. The treatment for AVMs aimed to minimize the risk of blood vessel rupture and resect lesions. Currently, treatment approaches mainly include microsurgical resection, endovascular embolization, and stereotactic radiosurgery [1, 2, 26].

Microsurgical resection
Microsurgical resection is still a primary or "gold-standard" method for the authoritative management of low-grade AVMs [25]. Its main advantages lie in the high rate of complete elimination, the immediate elimination of bleeding risk, and its long-term stability, thus reducing morbidity and mortality [34].
Johannes et al reported that a consecutive series of 288 cases underwent microsurgical resection. And microsurgical resection of a cerebral AVM has a high cure rate [29]. Although microsurgery is currently the mainstream surgical method for the treatment of AVMs, it also has its shortcomings, such as large trauma, long recovery time, and the risk of related neurological defects. Endovascular embolization is often used to improve the safety of microsurgery. The main objectives of preoperative embolization include occlusion of the supplying arteries, reduction of blood ow or malformation of vascular mass volume to allow safer surgical excision, and management of those high-risk angiography features, including aneurysm located on supplying arteries and malformation [28]. Intraoperative angiography has also been used to guide AVM surgical treatment. The main purpose of DSA during the operation was to verify whether the deformed vascular mass was completely removed, while the main purpose of uorescein or indocyanine green angiography during the operation was to show AVM vascular architecture. Therefore, the use of these surgical auxiliary means may improve patients' selection, reduce surgical complications, and promote patients' recovery, although these advantages are di cult to be de nitively determined in clinical studies.

Endovascular embolization
Endovascular embolization has been generally recognized as a safe and effective means for the AVMs [18]. It is generally accepted that endovascular embolization is also an effective adjunctive treatment in a variety of clinical situations to facilitate complete surgical excision and reduce the risk of intra-operative bleeding [8,33]. In some cases, the healing effect of AVMs is achieved by embolization alone, which is helpful in the application of the healing embolization strategy [12]. This is corroborated in the study by Adam et al [27]. They drew the conclusion that only 15% of patients achieved safely complete angiographic obliteration of the bAVM with embolization alone. However, our outcomes of complete bAVM obliteration are far superior to their study. They speculate that proximal occlusion of feeding arteries may be associated with recurrence. Another indication for endovascular therapy is as a substitute for microsurgery or stereotactic radiosurgery [36]. In this study, endovascular embolization helps to reduce AVMs volume or occlusion of high-risk characteristics, such as ruptured aneurysms in or around the deformed vascular cluster, and nal treatment of remaining AVMs. Finally, embolization has been used as palliative care to reduce vascular irregular blood ow to malformed vascular masses, thereby improving the underlying symptoms associated with blood theft.

Hybrid operation
Recently, multimodality treatment, especially hybrid surgery, has been paid more and more attention as an effective treatment of intracranial AVM [8,9]. Preoperative partial embolization of the malformation can assist surgical positioning, reduce blood ow to the malformation, and reduce intraoperative bleeding risk and surgical di culty [5]. Intraoperative angiography can detect the residual malformation immediately for one-stop resection, greatly reducing the residual rate of postoperative malformation and the risk of postoperative rebleeding [16,17,21,31]. Despite the lack of large sample evidence, multiple single-center randomized controlled studies have reported that the application of a hybrid operating room provides more satisfactory therapeutic effects compared with traditional surgery. However, for grade III or above, especially for complex bAVM lesions located in functional areas, simultaneous intraoperative angiography or embolization combined microsurgical resection under a hybrid operating room can achieve a better therapeutic effect and long-term prognosis, as well as a good functional protective effect.
Few reports were addressing the surgical e cacy between several above treatments. Procedure-related complications were higher in the hybrid group in comparison with the embolization alone group in our study, but the results showed no signi cant difference between the hybrid operation and embolization alone groups in any of the favorable clinical outcomes. These results are consistent with our results. Wang et al. found that better long-term outcomes in 92.5% of unruptured AVM patients compared with ruptured AVMs in long-term follow-up [32]. These results support our nding that the immediate neurological de cits are usually transient and that many patients recover over time. Immediate postoperative radiological ndings demonstrated complete obliteration in 99.4% of patients in the hybrid surgery group, whereas 38.3% in the embolization alone group. Wen et al. also found that the hybrid operation presented distinct advantages in cure rates, especially for complex AVMs [33]. In our series, the cured rate of the embolization alone group was lower than that in other studies on the safety of embolization with large case series. Among the embolization alone group, disorders were located in functional areas in a large percentage of patients. This selection bias for embolization might be the cause of a lower cure rate in the embolization alone group. In our study, the removal of AVM in several patients was supposed to be completed safely through the hybrid operation in our cases, though with high-grade AVM lesions, which was similar to Wen's results [33]. Furthermore, the mortality rate in the hybrid group was much lower than that in the embolization alone group, although this difference was not found to be statistically signi cant. Wang et al. reported lower operation-related morbidity in their series of low-grade AVMs. They attributed the low morbidity to several reasons [32]. The most important target is embolizing the feeding arteries to reduce intraoperative bleeding rather than the AVM nidus. We also believed that it was important to follow this strategy, particularly for high SM grade and complex AVMs.

Limitations
Admittedly, our research also has some defects. First, the main limitation of our study is a retrospective analysis, single-center, non-randomized study design, and sample size of patients with cerebral arteriovenous malformation small, which may result in patient selection and registration of bias. Secondly, the clinical results are carefully evaluated through medical record review, but they are prone to deviation without blind evaluation and document errors. Finally, some patients were lost to follow-up, although the low rate did not in uence nal results.

Conclusion
One-stop hybrid operation combined embolization and microsurgical resection could be a safe and effective method of intervention for the AVMs. It can be used for effective and complete occlusion of lowgrade or high-grade and complex arteriovenous malformations. Certainly, further follow-up is needed to determine the long-term effects after operation.

Declarations
Funding: The study was supported by the Fundamental Research Funds for the Central Universities of China (No. 2042020kf0166).
Con ict of interest: The authors declare no con ict of interest.
Availability of data and material: Restrictions apply to the availability of these data due to privacy restrictions.   Flow diagram for the process of study screening and assignment of the study patients.

Figure 2
Page 19/20 Distribution of SM Grade in bAVM patients.