Risk Assessment of Hemorrhage of Posterior Inferior Cerebellar Artery Aneurysms in Posterior Fossa Arteriovenous Malformations

Risk Assessment of Hemorrhage of Posterior Inferior Cerebellar Artery Aneurysms in Posterior... Abstract BACKGROUND Posterior fossa arteriovenous malformations (AVMs) are associated with increased risk of rupture and severe consequences from such rupture. The hemorrhagic risk of prenidal aneurysms (anr) on the posterior inferior cerebellar artery (PICA) may exceed that of the AVM in posterior fossa AVMs fed by PICA (PICA-AVM). OBJECTIVE To characterize the relative risks of aneurysm and AVM hemorrhage in patients with posterior fossa AVMs. METHODS We retrospectively reviewed patients diagnosed with AVM. Patients with posterior fossa AVMs were divided into 3 groups: PICA-AVM with prenidal aneurysm (PICA-AVM-anr group), PICA-AVM without prenidal aneurysm (PICA-AVM group), and AVMs without PICA feeder with/without aneurysm (AVM-only group). Patient and lesion characteristics and treatment outcomes were compared. ANOVA and chi squared tests were used for statistical analyses. RESULTS Our cohort included 85 patients. Mean age was 45.3 ± 18.1 yr, with 43(50.6%) female patients. Fifty-one patients (60.0%) had hemorrhagic presentation, and 27 (31.8%) experienced acute hydrocephalus. Patients in the PICA-AVM-anr group (n = 11) were more likely to present with aneurysmal subarachnoid hemorrhage (SAH; P = .005) and less likely to have AVM rupture (P < .001). Ten (90.9%) patients presented with hemorrhage, 6 (60.0%) of which resulted from aneurysm rupture. Of these 6, 5 (83.3%) had acute hydrocephalus. No patients with AVM rupture had hydrocephalus. Eight (72.7%) received aneurysm treatment prior to AVM treatment. There were no significant differences in post-treatment outcomes dependent on treatment order. CONCLUSION In addition to relatively higher risk of AVM rupture from infratentorial location and prenidal aneurysm, a higher risk of aneurysm rupture rather than AVM rupture was observed in patients with PICA-AVM-anr complex. Arteriovenous malformation (AVM), Feeding artery, Posterior fossa, Posterior inferior cerebellar artery (PICA), Prenidal aneurysm, Feeding ABBREVIATIONS ABBREVIATIONS anr aneurysm AVM arteriovenous malformation CT computed tomography ICH intracerebral hemorrhage IVH intraventricular hemorrhage MRI magnetic resonance imaging mRS modified Rankin score PICA posterior inferior cerebellar artery SAH subarachnoid hemorrhage Brain arteriovenous malformations (AVMs) commonly manifest with hemorrhage as the initial presentation.1-3 Studies on the natural history of AVMs suggest that the risk of hemorrhagic presentation is 30% to 82%, with a subsequent annual risk of 1.9% to 4.6% in untreated lesions.1,2,4-7 The risk is increased with certain AVM characteristics, including smaller size, associated aneurysms, deep venous drainage, and location.8-10 Posterior fossa AVMs, specifically, have been shown to be more prone to rupture than supratentorial AVMs.11-15 Meta-analysis of patients with posterior fossa AVMs demonstrated 84% risk of hemorrhagic presentation,14 with subsequent annual rupture rate ranging from 7.5% to 11.6% in the first 5 yr.16-18 Therefore, timely diagnosis and treatment of infratentorial lesions are imperative. Surgical treatment risk of posterior fossa AVMs is higher than that of supratentorial AVMs, both due to technical difficulty and postoperative risk. The technical challenge involves the preservation of nearby cranial nerves and brainstem nuclei, while navigating narrow surgical corridors.19-21 Moreover, limited posterior fossa space increases the risk of postsurgical hydrocephalus and poor functional outcomes.22,23 In association with feeding artery aneurysms, the selection of treatment strategy is further complicated by the comparative risk of hemorrhage between the AVM and aneurysm. Feeding artery aneurysms are independent predictors of AVM rupture and subsequent poor outcome, with a 6% attributable risk to hemorrhagic presentation.15,24-27 In some cases, the feeding artery aneurysm is the source of the initial bleed,28-30 and exceedingly high risk of prenidal aneurysmal rupture has been reported in aneurysms on the posterior inferior cerebellar artery (PICA) feeding artery of posterior fossa AVMs.30 In light of these considerations, significant concerns arise regarding treatment priorities between AVMs supplied by PICA (PICA-AVM) and associated prenidal feeding artery aneurysms (PICA-AVM-anr). Better understanding of the risk of hemorrhage and postoperative outcomes of these lesions is critical in making informed decisions. The current literature is limited regarding PICA-AVM combined with feeding artery aneurysms, with only 30 cases reported to date.31 The goal of this study, therefore, is to focus on the subset of patients with this combination, and to compare the hemorrhagic risk and post-treatment outcomes of each. We hypothesize that despite an increased risk of rupture associated with infratentorial location and prenidal aneurysms, the risk of AVM hemorrhage may be overshadowed by the exceeding risk of aneurysmal hemorrhage in patients with PICA-AVM-anr. METHODS Study Population This study is designed as a retrospective cohort study. For this IRB-approved study, we retrospectively reviewed the electronic medical records of 683 patients diagnosed with brain AVMs evaluated at our institution between January 1990 and December 2013. Patient data were collected via IRB-approved database, and consent was exempted given the retrospective nature of this study. Patients with hereditary hemorrhagic telangiectasia, missing baseline information, or lost to follow-up were excluded. The patient selection flowchart is depicted in Figure 1. We only included patients with posterior fossa (infratentorial) AVMs, and the study cohort was divided into 3 groups: PICA-AVM with prenidal aneurysm (PICA-AVM-anr group), PICA-AVM without prenidal aneurysm (PICA-AVM group), and AVMs without PICA feeding artery with/without aneurysm (AVM-only group). Prenidal aneurysm in the PICA-AVM-anr group included both distal feeding aneurysms and those located at the PICA origin. FIGURE 1. View largeDownload slide Flowchart of patient selection process. FIGURE 1. View largeDownload slide Flowchart of patient selection process. Definition of Variable and Outcome Demographic information included sex, age, and race. Age was defined as age at the time of AVM diagnosis. Type of presenting hemorrhage was determined via direct image review or documentation, with AVM rupture defined by intracerebral hemorrhage (ICH), subarachnoid hemorrhage (SAH), and intraventricular hemorrhage (IVH). The source of hemorrhage, if present, was specified as AVM or aneurysm based on image interpretations by both attending neurosurgeon and radiologist. In general, AVM rupture usually presents with ICHs that are largely intraparenchymal and correspond with the location of the AVM. In contrast, aneurysmal hemorrhage predominantly presents with SAH with or without IVH and correspond with the location of the aneurysm. A representative case is presented in Figure 2 to depict the computed tomography (CT) appearance of hemorrhagic presentation with AVM as the primary source of bleed. This case can be contrasted with Figure 3 (PICA-AVM-anr with aneurysm as primary source of bleeding) to distinguish the different appearance on CT. Presence of acute hydrocephalus upon presentation and baseline symptoms (headaches, nausea/vomiting, and dizziness) were noted. AVM lesion characteristics, including nidus size, deep venous drainage, and location, were assessed via baseline angiograms, CT images, and magnetic resonance imaging (MRI). Spetzler–Martin grading was then determined using these 3 parameters. Other angiographic features, such as presence of intranidal aneurysms or other feeding arteries, were also included in the description of each AVM. Morphology and size for each aneurysm were described. FIGURE 2. View largeDownload slide In order to illustrate the difference of hemorrhagic pattern between aneurysmal rupture and AVM rupture, we provided an additional case of hemorrhagic presentation with ruptured AVM in a patient with PICA-AVM (without prenidal aneurysm) to contrast with the CT appearance of Figure 3. A, Axial CT showing predominant left cerebellar intraparenchymal hemorrhage with heterogeneous high-density ICH. B, ICH breaking into the fourth ventricle forming IVH. C, Although with SAH and IVH, which is similar to Figure 3, the evolution of ICH to ventricular breakthrough forming into IVH/SAH is representative of AVM rupture, which is distinctly different from an aneurysmal bleeding demonstrated in Figure 3. D, Left vertebral injection showing a large left AVM with SCA and PICA feeders. FIGURE 2. View largeDownload slide In order to illustrate the difference of hemorrhagic pattern between aneurysmal rupture and AVM rupture, we provided an additional case of hemorrhagic presentation with ruptured AVM in a patient with PICA-AVM (without prenidal aneurysm) to contrast with the CT appearance of Figure 3. A, Axial CT showing predominant left cerebellar intraparenchymal hemorrhage with heterogeneous high-density ICH. B, ICH breaking into the fourth ventricle forming IVH. C, Although with SAH and IVH, which is similar to Figure 3, the evolution of ICH to ventricular breakthrough forming into IVH/SAH is representative of AVM rupture, which is distinctly different from an aneurysmal bleeding demonstrated in Figure 3. D, Left vertebral injection showing a large left AVM with SCA and PICA feeders. FIGURE 3. View largeDownload slide A, Preoperative axial CT showing hemorrhage in the fourth ventricle. B, Preoperative CT showing IVH in the lateral ventricles. C, Anterior-posterior (AP) view of a left vertebral injection on digital subtraction angiography (DSA) demonstrating both the prenidal aneurysm (solid black arrow) and the AVM nidus (dashed black arrow). D, Lateral view of the aneurysm and AVM. E, AP view on DSA showing obliteration of both the aneurysm and AVM following aneurysm embolization and AVM resection. F, Postoperative lateral view DSA. FIGURE 3. View largeDownload slide A, Preoperative axial CT showing hemorrhage in the fourth ventricle. B, Preoperative CT showing IVH in the lateral ventricles. C, Anterior-posterior (AP) view of a left vertebral injection on digital subtraction angiography (DSA) demonstrating both the prenidal aneurysm (solid black arrow) and the AVM nidus (dashed black arrow). D, Lateral view of the aneurysm and AVM. E, AP view on DSA showing obliteration of both the aneurysm and AVM following aneurysm embolization and AVM resection. F, Postoperative lateral view DSA. Functional status at baseline and at last follow-up was assessed using modified Rankin scale (mRS).32-34 Follow-up duration was defined as the interval between AVM diagnosis and last follow-up. AVM obliteration was determined using angiogram, CT image, or MRI at last follow-up. Symptoms (headaches, nausea/vomiting, and dizziness) at last follow-up were determined. Statistical Analysis Patient demographic and angiographic characteristics were compared across different groups. ANOVA and χ2 test were used to analyze continuous and categorical variables respectively. Data analysis was performed using R Statistical Software (Version 3.1.1, 2013, Vienna, Austria). Statistical significance was defined as P < .05. All P values were reported as 2-sided. RESULTS Study Cohort Characteristics A total of 93 patients with posterior fossa AVMs were retrieved from 683 patients. After applying our inclusion and exclusion criteria, our cohort consisted of 85 patients. Mean age of all patients was 45.3 ± 18.1 yr, with 43 (50.6%) female patients. Fifty-one patients (60.0%) had hemorrhagic presentation, and 27 (31.8%) patients experienced acute hydrocephalus. Among the other 34 patients without hemorrhagic presentation, 5 were incidentally diagnosed without any presenting symptoms, and the rest of the patients (n = 29) presented with nonhemorrhagic symptoms such as seizures, persistent headaches, visual disturbances, or cerebellar symptoms that prompted the patient for a neurological workup leading to the discovery of the AVM. Within our study cohort, there were 11 (12.9%) patients in the PICA-AVM-anr group, 32 (37.6%) patients in the PICA-AVM group, and 42 (49.4%) patients in the AVM-only group. Age, gender, and race were similarly distributed across the 3 groups (P = .603, .453, and .287, respectively). There was no significant difference in the proportion of patients presenting with ICH among the 3 groups (P = .191). However, the PICA-AVM and AVM-only groups had significantly fewer patients with presenting SAH than the PICA-AVM-anr group (P = .005). The PICA-AVM and AVM-only groups also showed a trend of fewer IVH than the PICA-AVM-anr group (P = .061). In the PICA-AVM-anr group, 36.4% of patients presented with AVM rupture, compared to 43.8% in the PICA-AVM group and 57.1% in the AVM-only group. For other patients without AVM rupture at presentation, only 1 patient (9.1%) in the PICA-AVM-anr group had nonhemorrhagic presentation, whereas the majority presented with aneurysmal rupture (n = 6, 54.5%). This is in contrast with a 56.3% and 33.3% nonhemorrhagic presentation in the PICA-AVM and AVM-only groups, and only 7.1% aneurysmal rupture in group 3 (P < .001). Despite a higher risk of hemorrhagic presentation in the PICA-AVM-anr group, patients in all 3 groups were equally likely to develop acute hydrocephalus (P = .850). Patients in the PICA-AVM-anr group were more likely to receive surgery ± embolization of the AVM, whereas the patients in the PICA-AVM and AVM-only groups were more likely to receive radiation ± embolization of the AVM. More patients in the PICA-AVM-anr group received embolization as the only treatment modality in regard to the AVM compared to patients in the other 2 groups (P = .006). A description of the comparison is presented in Table 1. TABLE 1. Comparison of Patient and Lesion Characteristics and Treatment Selection Between Different Groups   Total  PICA-AVM-anr  PICA-AVM  AVM-only    Parameters  (n = 85)  (n = 11)  (n = 32)  (n = 42)  P value  Age, mean (sd)  45.25 (18.15)  55.94 (6.89)  39.84 (16.24)  46.58 (19.98)  .603  Gender, male, n (%)  42 (49.4)  6 (54.5)  13 (40.6)  23 (54.8)  .453  Race          .287   White, n (%)  53 (62.4)  5 (45.5)  19 (59.4)  29 (69.0)     Black, n (%)  26 (30.6)  6 (54.5)  11 (34.4)  9 (21.4)     Other, n (%)  6 (7.1)  0 (0.0)  2 (6.3)  4 (9.5)    Spetzler–Martin grades, n (%)          .031*   Grade 1  17 (20.0)  5 (45.5)  6 (18.8)  6 (14.3)     Grade 2  31 (36.5)  4 (36.4)  16 (50.0)  11 (26.2)     Grade 3  30 (35.3)  1 (9.1)  9 (28.1)  20 (47.6)     Grade 4  5 (5.9)  0 (0.0)  1 (3.1)  4 (9.5)     Grade 5  2 (2.4)  1 (9.1)  0 (0.0)  1 (2.4)    AVM max size (cm), mean (sd)  2.34 (1.45)  2.31 (1.39)  2.75 (1.17)  2.03 (1.57)  .185   ICH  40 (47.1)  6 (54.5)  11 (34.4)  23 (54.8)  .191   SAH  15 (17.6)  6 (54.5)  5 (15.6)  4 (9.5)  .005   IVH  27 (31.8)  7 (63.6)  9 (28.1)  11 (26.2)  .061  AVM or aneurysmal rupture          <.001   Aneurysm, n (%)  9 (10.6)  6 (54.5)  0 (0.0)  3 (7.1)     AVM, n (%)  42 (49.4)  4 (36.4)  14 (43.8)  24 (57.1)     None, n (%)  34 (40.0)  1 (9.1)  18 (56.3)  15 (35.7)    Acute hydrocephalus  27 (31.8)  4 (36.4)  9 (28.1)  14 (33.3)  .850  Pretreatment mRS, mean (sd)  1.60 (1.10)  1.45 (0.99)  1.69 (1.21)  1.57 (1.02)  .956  PICA aneurysm  11 (12.9)  11 (100.0)  0 (0.0)  0 (0.0)  <.001  PICA ANR treatment before AVM  8 (9.4)  8 (72.7)  0 (0.0)  0 (0.0)  <.001  AVM treatment, n (%)          .006   Surgery ± embolization  14 (16.5)  4 (36.4)  7 (21.9)  3 (7.1)     Radiation ± embolization  39 (45.9)  1 (9.1)  16 (50.0)  22 (52.4)     Surgery + radiation  2 (2.4)  0 (0.0)  0 (0.0)  2 (4.8)     Embolization  7 (8.2)  3 (27.3)  3 (9.4)  1 (2.4)     Observation  23 (27.1)  3 (27.3)  6 (18.8)  14 (33.3)    AVM obliteration          .421   Yes, n (%)  29 (34.1)  6 (54.5)  10 (31.3)  13 (31.0)     No, n (%)  41 (48.2)  5 (45.5)  16 (50.0)  20 (47.6)     Unknown, n (%)  15 (17.6)  0 (0.0)  6 (18.8)  9 (21.4)      Total  PICA-AVM-anr  PICA-AVM  AVM-only    Parameters  (n = 85)  (n = 11)  (n = 32)  (n = 42)  P value  Age, mean (sd)  45.25 (18.15)  55.94 (6.89)  39.84 (16.24)  46.58 (19.98)  .603  Gender, male, n (%)  42 (49.4)  6 (54.5)  13 (40.6)  23 (54.8)  .453  Race          .287   White, n (%)  53 (62.4)  5 (45.5)  19 (59.4)  29 (69.0)     Black, n (%)  26 (30.6)  6 (54.5)  11 (34.4)  9 (21.4)     Other, n (%)  6 (7.1)  0 (0.0)  2 (6.3)  4 (9.5)    Spetzler–Martin grades, n (%)          .031*   Grade 1  17 (20.0)  5 (45.5)  6 (18.8)  6 (14.3)     Grade 2  31 (36.5)  4 (36.4)  16 (50.0)  11 (26.2)     Grade 3  30 (35.3)  1 (9.1)  9 (28.1)  20 (47.6)     Grade 4  5 (5.9)  0 (0.0)  1 (3.1)  4 (9.5)     Grade 5  2 (2.4)  1 (9.1)  0 (0.0)  1 (2.4)    AVM max size (cm), mean (sd)  2.34 (1.45)  2.31 (1.39)  2.75 (1.17)  2.03 (1.57)  .185   ICH  40 (47.1)  6 (54.5)  11 (34.4)  23 (54.8)  .191   SAH  15 (17.6)  6 (54.5)  5 (15.6)  4 (9.5)  .005   IVH  27 (31.8)  7 (63.6)  9 (28.1)  11 (26.2)  .061  AVM or aneurysmal rupture          <.001   Aneurysm, n (%)  9 (10.6)  6 (54.5)  0 (0.0)  3 (7.1)     AVM, n (%)  42 (49.4)  4 (36.4)  14 (43.8)  24 (57.1)     None, n (%)  34 (40.0)  1 (9.1)  18 (56.3)  15 (35.7)    Acute hydrocephalus  27 (31.8)  4 (36.4)  9 (28.1)  14 (33.3)  .850  Pretreatment mRS, mean (sd)  1.60 (1.10)  1.45 (0.99)  1.69 (1.21)  1.57 (1.02)  .956  PICA aneurysm  11 (12.9)  11 (100.0)  0 (0.0)  0 (0.0)  <.001  PICA ANR treatment before AVM  8 (9.4)  8 (72.7)  0 (0.0)  0 (0.0)  <.001  AVM treatment, n (%)          .006   Surgery ± embolization  14 (16.5)  4 (36.4)  7 (21.9)  3 (7.1)     Radiation ± embolization  39 (45.9)  1 (9.1)  16 (50.0)  22 (52.4)     Surgery + radiation  2 (2.4)  0 (0.0)  0 (0.0)  2 (4.8)     Embolization  7 (8.2)  3 (27.3)  3 (9.4)  1 (2.4)     Observation  23 (27.1)  3 (27.3)  6 (18.8)  14 (33.3)    AVM obliteration          .421   Yes, n (%)  29 (34.1)  6 (54.5)  10 (31.3)  13 (31.0)     No, n (%)  41 (48.2)  5 (45.5)  16 (50.0)  20 (47.6)     Unknown, n (%)  15 (17.6)  0 (0.0)  6 (18.8)  9 (21.4)    *Significant variables (P < .050) View Large Subcohort of PICA-AVM With Feeder Aneurysm There were 11 patients with PICA-AVM and prenidal PICA-feeding artery aneurysms, and detailed patient characteristics are summarized in Table 2. Mean age for these patients with PICA-AVM-anr was 55.9 ± 18.5 yr, with 6 (54.5%) male patients. There were 5 patients with Spetzler–Martin grade I AVM (45.5%), 4 patients with grade II (36.4%), and 1 patient each for grades IV and V (9.1%). None of the patients harbored intranidal aneurysms in the AVM. Ten (90.9%) patients had a hemorrhagic presentation, and 6 (60.0%) of them resulted from aneurysm rupture. Of those with aneurysm rupture, 5 (45.5%) presented with acute hydrocephalus, and all patients with acute hydrocephalus had IVH. No patient with AVM rupture presented with hydrocephalus. All patients who presented with ruptured feeder aneurysm had IVH. Only one patient (9.1%) had a saccular aneurysm. Other patients either had fusiform or irregularly shaped aneurysms. Eight patients (72.7%) had the aneurysm treated before the AVM, of which 5 (62.5%) were treated via embolization and 3 (37.5%) by microsurgical clipping. Nine out of the 11 patients (81.8%) received AVM treatment. Three (27.3%) received surgery only, 3 (27.3%) received embolization only, 1 (9.1%) received radiation only, 1 (9.1%) received surgery and embolization, 1 (9.1%) received radiation and embolization, and 2 (18.2%) were observed without treatment. Complete AVM obliteration was achieved in 6 (54.5%) patients. One patient (9.1%) received treatment for neither the aneurysm nor the AVM. The mean follow-up duration was 52.8 ± 63.5 mo. The average mRS for surgical patients compared to nonsurgical patients is 1.4 vs 1.3. TABLE 2. Characteristics of Patients with PICA Arteriovenous Malformation and PICA Aneurysm (PICA-AVM-anr)   Baseline characteristics  Lesion characteristics    Demographics  Presentation  AVM characteristic  Aneurysm characteristics  Treatment  Follow-up  Pt  Age sex  Race  Symptoms  Hemorrhage  Ruptured lesion  Hydro  Size(mm)/location  DV  Grade  Intranidal anr  Shape  Size (mm)  First  anr  AVM  Length (month)  AVM oblit  Hemorrhage  Symptoms  1  69F  W  H/IB/D/N/BV  +  anr  –  11/right CH  +  II  –  –  2 × 7  AVM  –  S  23  +  _  G  2  49F  W  H/S  +  anr  –  10/left CH  +  II  –  –  6.5  anr  E  E  6  –  –  H/G/HR  3  56M  AA  HR/D/H/IB/BV  –  –  –  29/V  +  II  –  –  –  –  –  –  3  –  –  HR/D/H/BV/G  4  52F  AA  H/D/N/V/IB  +  anr  +  20/V  –  I  –  wide neck  7.8  anr  E  E  2  –  ±  N/V  5  58M  W  H/D  +  AVM  –  15/right CH  –  I  –  +  2.5  anr  C  –  46  +  –  Sensory                        –  4.5                6  58M  AA  H/IB/W  +  AVM  –  34/right CH  +  III  –  –  7 × 8  AVM  –  ER  28  –  –  –  7  43M  AA  H/N/V/IC  +  anr  +  60/Left CH  +  V  –  Irregular  9  anr  E  E  91  –  +  H/W  8  60F  AA  H  +  AVM  –  15/V  +  II  –  –  –  anr  C  R  24  +  –  H/G/sensory  9  60F  AA  H/N/V/IB  +  anr  +  12.5/right CH  –  I  –  –  3  anr  E  ES  35  +  –  –  10  61M  W  H  +  anr  +  28/left CH  –  I  –  –  3.5  anr  E  S  26  +  –  dysphagia  11  48M  W  H/N  +  AVM  –  20/right CH  –  I  –  –  –  anr  C  S  144  +  –  G/BV    Baseline characteristics  Lesion characteristics    Demographics  Presentation  AVM characteristic  Aneurysm characteristics  Treatment  Follow-up  Pt  Age sex  Race  Symptoms  Hemorrhage  Ruptured lesion  Hydro  Size(mm)/location  DV  Grade  Intranidal anr  Shape  Size (mm)  First  anr  AVM  Length (month)  AVM oblit  Hemorrhage  Symptoms  1  69F  W  H/IB/D/N/BV  +  anr  –  11/right CH  +  II  –  –  2 × 7  AVM  –  S  23  +  _  G  2  49F  W  H/S  +  anr  –  10/left CH  +  II  –  –  6.5  anr  E  E  6  –  –  H/G/HR  3  56M  AA  HR/D/H/IB/BV  –  –  –  29/V  +  II  –  –  –  –  –  –  3  –  –  HR/D/H/BV/G  4  52F  AA  H/D/N/V/IB  +  anr  +  20/V  –  I  –  wide neck  7.8  anr  E  E  2  –  ±  N/V  5  58M  W  H/D  +  AVM  –  15/right CH  –  I  –  +  2.5  anr  C  –  46  +  –  Sensory                        –  4.5                6  58M  AA  H/IB/W  +  AVM  –  34/right CH  +  III  –  –  7 × 8  AVM  –  ER  28  –  –  –  7  43M  AA  H/N/V/IC  +  anr  +  60/Left CH  +  V  –  Irregular  9  anr  E  E  91  –  +  H/W  8  60F  AA  H  +  AVM  –  15/V  +  II  –  –  –  anr  C  R  24  +  –  H/G/sensory  9  60F  AA  H/N/V/IB  +  anr  +  12.5/right CH  –  I  –  –  3  anr  E  ES  35  +  –  –  10  61M  W  H  +  anr  +  28/left CH  –  I  –  –  3.5  anr  E  S  26  +  –  dysphagia  11  48M  W  H/N  +  AVM  –  20/right CH  –  I  –  –  –  anr  C  S  144  +  –  G/BV  Age/sex: M: Male, F: Female; Race: W: White, AA: African American; Symptoms: H: Headache, G: gait disturbance/imbalance, D: dizziness, N: nausea, V: vomiting, BV: blurry vision, S: seizure, HR: hearing disturbances, W: weakness, IC: incontinence; Location: CH: cerebellar hemisphere, V: vermis; Treatment: E: embolization, R: radiation, S: surgery, C: microsurgical clipping. View Large All patients presented with headaches. The next most common presenting symptom was imbalance (45.5%), followed by dizziness and nausea (36.4%). Other less common symptoms included blurry vision (18.2%), seizure (9.1%), hearing disturbances (9.1%), weakness (9.1%), and incontinence (9.1%). At last follow-up, the most common symptoms were headaches (36.4%) and imbalance (27.3%). Other symptoms included hearing disturbance, blurry vision, weakness, dysphagia, gait imbalance, and sensory-related symptoms. Case Presentation The patient was a 61-yr-old male who presented with severe headache. Head CT showed SAH with IVH and acute hydrocephalus (Figures 3A and 3B). Angiography revealed a left cerebellar AVM with the nidus measuring 2.8 × 1.0 × 1.0 cm, supplied by multiple distal branches of the left PICA and SCA (Figures 3C and 3D). Venous drainage was superficial via a left tentorial vein to the left transverse sinus. Source of hemorrhage was determined as a ruptured flow-related aneurysm arising from a distal branch of the left PICA. Embolization on the same day targeted the 2 feeding artery branches from the left PICA to the AVM including the parent branch of the PICA aneurysm, resulting in a 30% to 40% nidal volume reduction. Repeat angiography after recovery at 3-mo showed a 2.1 × 1.1 cm AVM in the left cerebellar hemisphere. AVM resection via a suboccipital approach was performed with intraoperative angiography for confirmation AVM obliteration. At 3-mo follow-up, the patient was almost back to baseline, with some residual dysphagia. Repeat angiography at 1 yr showed no residual or recurrent AVM or aneurysm (Figures 3E and 3F). DISCUSSION AVMs with associated feeding artery aneurysms are at increased risk of hemorrhage, although the relative risk of rupture of the AVM compared to the prenidal aneurysm is unclear. The elucidation of the comparative risk between these 2 lesions is critical for treatment decision making, as the core aim of an aneurysm-first or AVM-first treatment approach is to prevent subsequent hemorrhage during the interval between treatments. This study aims to determine this risk with a specific focus on PICA-AVM-anr complex, with a hypothesis derived from clinical impression that more cases of aneurysm rupture were seen compared to AVM rupture. Our study included 11 patients with PICA-AVM-anr complex among a total of 85 patients with posterior fossa AVMs, representing one of the largest series reported to date.30,31,35 Confirming our hypothesis, the majority of patients with both PICA feeding AVMs and PICA aneurysms (PICA-AVM-anr) presented with hemorrhage (90.9%), of which 60% were secondary to the aneurysm rupture. This is in concordance with a literature review of 27 PICA-AVM-anr cases by Kaptain and authors,30 in which 26 patients presented with hemorrhage, and 61.5% of all hemorrhaged cases were confirmed to present with aneurysmal bleeding, with only 7.7% of the hemorrhages attributable to AVM. Based on our results and previous reports that the risk of aneurysm rupture may overshadow AVM rupture, this unique disease combination may distinguish itself from other posterior fossa AVMs or AVM-anr complexes supplied by other feeding arteries, and deserves a thorough re-evaluation of best therapeutic approach to minimize overall risk of hemorrhage. There is currently no consensus on whether the AVM should be treated first or the associated feeding artery aneurysm. Demonstration of aneurysm recession following AVM resection in some reported cases, in particular cases with smaller prenidal aneurysms, suggests that AVMs should be treated before associated prenidal aneurysms in this cohort of patients. This can be explained by 2 theories: restoration of normal blood flow, which reduces aneurysmal dilatation, or flow reduction that promotes aneurysm sac thrombosis.35,36 However, the reduction in blood flow in feeder vessels following AVM resection may also increase the risk of subsequent hemorrhage via the normal perfusion pressure breakthrough theory, especially in larger or high-grade (grade 4 or 5) AVMs where blood-steal phenomenon is predominant. Normal perfusion pressure breakthrough theory suggests that the presence of AVM results in the chronic reduction in blood flow to normal tissue. As a result, the vessels supplying them dilate and lose their ability to autoregulate. Following AVM excision, the transient increase in perfusion to these vessels, coupled with their inability to vasoconstrict, predisposes the patient to further hemorrhage from the untreated feeder aneurysm.37,38 Therefore, for selected patients identified with high risk of aneurysm rupture, the aneurysm treatment should come before AVM resection. Existing studies on posterior fossa AVMs with feeder aneurysms also support this recommendation.28,29 It is unknown why PICA-AVM-anr bear a significantly higher risk of aneurysm rupture than other AVM-anr complexes; 1 possibility is that PICA-feeding artery aneurysms alone may have an overall higher risk of hemorrhage compared to other posterior circulation aneurysms regardless of existence of AVMs. From the results of the International Subarachnoid Aneurysm Trial, PICA feeding artery aneurysms accounted for 53.4% of all posterior fossa aneurysm ruptures. Interestingly, the second highest risk of ruptured presentation was basilar apex aneurysms (29.3%), which is with minimal association with feeding of AVM. In contrast, for superior cerebellar arteries or posterior cerebral arteries that have the potential to feed AVMs, only 9 cases (0.4%) were noted to have ruptured in 2143 cases.39 It is therefore reasonable to assume that the higher risk of aneurysmal hemorrhage in PICA-AVM-anr is likely to be related intrinsically to PICA itself, either due to particular anatomic reason or hemodynamic characteristics, as opposed to the impact from the combination of PICA-AVM. Future studies are warranted to further elucidate the underlying cause of this phenomenon. In our study, 8 out of 10 patients (80%) who underwent treatment had received aneurysm treatment prior to AVM treatment. Two patients had recurrent hemorrhage after aneurysm treatment. The first patient (patient 4; Table 2) was a 52-yr-old female with a grade I AVM and a 7.8 mm PICA aneurysm that received coil embolization. She experienced mild acute intraparenchymal hemorrhage 4 d after aneurysm treatment. The second patient (patient 7; Table 2) was a 43-yr-old male with a Spetzler–Martin grade V AVM and a 9 mm aneurysm that was embolized. He had recurrent hemorrhage 3 yr after initial aneurysm embolization. Angiography at the time showed persistent AVM without significant feeding artery aneurysms as the source of bleed. The patient continued to experience weakness and headaches until last follow-up. These results suggest that despite aneurysm treatment, there is still risk for recurrent hemorrhage. In 1 patient (patient 5; Table 2), the AVM spontaneously thrombosed on angiography at 6 mo after aneurysm treatment. This supports the theory that treating feeder aneurysms prior to treating AVM may reduce the flow to the associated AVM. This may result in a decrease in AVM size or, as in this case, lead to thrombosis. However, it is important to note that spontaneous thrombosis following embolization of a feeder aneurysm should not be the expectation following aneurysm treatment. Definitive treatment of the AVM should still be sought. Of the 2 patients who received AVM treatment first (patients 1 and 6; Table 2), 1 was asymptomatic at 6-mo follow-up, while the other experienced elevated intracranial pressure and residual imbalance. These findings were not significantly better or worse than follow-up symptoms of those who received aneurysm treatment first. This suggests that there is no obvious difference in functional outcome between those with aneurysm treated first and those with AVM treated first. Within this population, acute hydrocephalus arising after hemorrhage is potentially catastrophic due to proximity of the fourth ventricle and cerebral aqueduct. Therefore, correlation of hydrocephalus risk with AVM and aneurysm rupture is essential in formulating treatment plans. Our cohort included 4 patients (36.4%) who presented with acute hydrocephalus, all of whom were found to have IVH. All 4 patients hemorrhaged because of PICA feeding artery aneurysm rupture. No patients who presented with AVM rupture experienced acute hydrocephalus. Furthermore, all patients with aneurysm rupture were found to have IVH, including those who did not meet criteria for acute hydrocephalus. This suggests that PICA aneurysm rupture may be correlated with a higher complication rate of hydrocephalus than PICA-AVM rupture—a likely consequence of development of IVH, which predisposes patients to acute hydrocephalus. There was no significant difference in residual symptoms at last follow-up. Limitations There are several study limitations that must be highlighted. As a single-institution study, sampling bias limits the generalizability of our results. Furthermore, there is also attrition bias due to the exclusion of patients who were lost to follow-up or had missing critical information. The small sample size makes it difficult to reach sufficient statistical power. Another potential limitation is the variability of treatment modality demonstrated in this study cohort. As treatment selection is determined by a shared decision-making process with patient preference having a significant weight in our institution, a variability of treatment was observed, especially for low-grade patients where surgical resection was regarded as the best treatment. However, this also provided us with the advantage of comparing disease characteristics by different treatment modalities with lower risk of selection bias. CONCLUSION In this report of a rare combination of PICA-AVM-anr, patients are more likely to present with hemorrhage from the associated PICA feeding artery aneurysm as opposed to the posterior fossa AVM. The concordance of our results with previous studies warrants a thorough evaluation of management strategy when addressing this disease combination. We suggest that a PICA-AVM-anr complex should be considered for an aneurysm-first strategy to reduce risk of aneurysm rupture. AVM treatment should follow promptly after aneurysm treatment in order to reduce the risk of hemorrhage from the AVM that is potentially attributable to disturbance of hemodynamics after prenidal aneurysm treatment. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Brown RD, Wiebers DO, Forbes G et al.   The natural history of unruptured intracranial arteriovenous malformations. J Neurosurg . 1988; 68( 3): 352- 357. Google Scholar CrossRef Search ADS PubMed  2. Fleetwood IG, Steinberg GK. Arteriovenous malformations. Lancet . 2002; 359( 9309): 863- 873. Google Scholar CrossRef Search ADS PubMed  3. Ondra SL, Troupp H, George ED, Schwab K. The natural history of symptomatic arteriovenous malformations of the brain: a 24-year follow-up assessment. J Neurosurg . 1990; 73( 3): 387- 391. Google Scholar CrossRef Search ADS PubMed  4. da Costa L, Wallace MC, Ter Brugge KG, O’Kelly C, Willinsky RA, Tymianski M. The natural history and predictive features of hemorrhage from brain arteriovenous malformations. Stroke . 2009; 40( 1): 100- 105. Google Scholar CrossRef Search ADS PubMed  5. Graf CJ, Perret GE, Torner JC. Bleeding from cerebral arteriovenous malformations as part of their natural history. J Neurosurg . 1983; 58( 3): 331- 337. Google Scholar CrossRef Search ADS PubMed  6. Gross BA, Du R. Natural history of cerebral arteriovenous malformations: a meta-analysis. J Neurosurg . 2013; 118( 2): 437- 443. Google Scholar CrossRef Search ADS PubMed  7. Hernesniemi JA, Dashti R, Juvela S, Väärt K, Niemelä M, Laakso A. Natural history of brain arteriovenous malformations: a long-term follow-up study of risk of hemorrhage in 238 patients. Neurosurgery . 2008; 63( 5): 829- 831. Google Scholar CrossRef Search ADS   8. Dashe JF. Predictors of hemorrhage in patients with untreated brain arteriovenous malformation. Neurology . 2007; 68( 7): 535; author reply 535. Google Scholar CrossRef Search ADS PubMed  9. Langer DJ, Lasner TM, Hurst RW, Flamm ES, Zager EL, King JT. Hypertension, small size, and deep venous drainage are associated with risk of hemorrhagic presentation of cerebral arteriovenous malformations. Neurosurgery . 1998; 42( 3): 481- 489. Google Scholar CrossRef Search ADS PubMed  10. Stefani MA, Porter PJ, terBrugge KG, Montanera W, Willinsky RA, Wallace MC. Angioarchitectural factors present in brain arteriovenous malformations associated with hemorrhagic presentation. Stroke . 2002; 33( 4): 920- 924. Google Scholar CrossRef Search ADS PubMed  11. Batjer H, Samson D. Arteriovenous malformations of the posterior fossa. Clinical presentation, diagnostic evaluation, and surgical treatment. J Neurosurg . 1986; 64( 6): 849- 856. Google Scholar CrossRef Search ADS PubMed  12. Brown RD, Wiebers DO, Torner JC, O’Fallon WM. Frequency of intracranial hemorrhage as a presenting symptom and subtype analysis: a population-based study of intracranial vascular malformations in Olmsted Country, Minnesota. J Neurosurg . 1996; 85( 1): 29- 32. Google Scholar CrossRef Search ADS PubMed  13. Drake CG, Friedman AH, Peerless SJ. Posterior fossa arteriovenous malformations. J Neurosurg . 1986; 64( 1): 1- 10. Google Scholar CrossRef Search ADS PubMed  14. Arnaout OM, Gross BA, Eddleman CS, Bendok BR, Getch CC, Batjer HH. Posterior fossa arteriovenous malformations. Neurosurg Focus . 2009; 26( 5): E12. Google Scholar CrossRef Search ADS PubMed  15. Khaw AV, Mohr JP, Sciacca RR et al.   Association of infratentorial brain arteriovenous malformations with hemorrhage at initial presentation. Stroke . 2004; 35( 3): 660- 663. Google Scholar CrossRef Search ADS PubMed  16. Kelly ME, Guzman R, Sinclair J et al.   Multimodality treatment of posterior fossa arteriovenous malformations. J Neurosurg . 2008; 108( 6): 1152- 1161. Google Scholar CrossRef Search ADS PubMed  17. Mine S, Hirai S, Ono J, Yamaura A. Risk factors for poor outcome of untreated arteriovenous malformation. J Clin Neurosci . 2000; 7( 6): 503- 506. Google Scholar CrossRef Search ADS PubMed  18. Yamada S, Takagi Y, Nozaki K, Kikuta K, Hashimoto N. Risk factors for subsequent hemorrhage in patients with cerebral arteriovenous malformations. J Neurosurg . 2007; 107( 5): 965- 972. Google Scholar CrossRef Search ADS PubMed  19. Drake CG. Cerebral arteriovenous malformations: considerations for and experience with surgical treatment in 166 cases. Clin Neurosurg . 1979; 26: 145- 208. Google Scholar CrossRef Search ADS PubMed  20. Neacsu A, Ciurea AV. General considerations on posterior fossa arteriovenous malformations (clinics, imaging and therapy). Actual concepts and literature review. J Med Life . 2010; 3( 1): 26- 35. Google Scholar PubMed  21. Solomon RA, Stein BM. Management of arteriovenous malformations of the brain stem. J Neurosurg . 1986; 64( 6): 857- 864. Google Scholar CrossRef Search ADS PubMed  22. Torné R, Rodríguez-Hernández A, Arikan F et al.   Posterior fossa arteriovenous malformations: Significance of higher incidence of bleeding and hydrocephalus. Clin Neurol Neurosurg . 2015; 134: 37- 43. Google Scholar CrossRef Search ADS PubMed  23. Yang W, Wang JY, Caplan JM et al.   Predictors of functional outcome following treatment of posterior fossa arteriovenous malformations. J Clin Neurosci . 2015; 22( 2): 357- 362. Google Scholar CrossRef Search ADS PubMed  24. da Costa L, Thines L, Dehdashti AR et al.   Management and clinical outcome of posterior fossa arteriovenous malformations: report on a single-centre 15-year experience. J Neurol Neurosurg Psychiatry . 2009; 80( 4): 376- 379. Google Scholar CrossRef Search ADS PubMed  25. Abla AA, Nelson J, Rutledge WC, Young WL, Kim H, Lawton MT. The natural history of AVM hemorrhage in the posterior fossa: comparison of hematoma volumes and neurological outcomes in patients with ruptured infra- and supratentorial AVMs. Neurosurg Focus . 2014; 37( 3): E6. Google Scholar CrossRef Search ADS PubMed  26. Stapf C, Mohr JP, Pile-Spellman J et al.   Concurrent arterial aneurysms in brain arteriovenous malformations with haemorrhagic presentation. J Neurol Neurosurg Psychiatry . 2002; 73( 3): 294- 298. Google Scholar CrossRef Search ADS PubMed  27. Thompson RC, Steinberg GK, Levy RP, Marks MP. The management of patients with arteriovenous malformations and associated intracranial aneurysms. Neurosurgery . 1998; 43( 2): 202- 211. Google Scholar CrossRef Search ADS PubMed  28. Kouznetsov E, Weill A, Ghostine JS, Gentric JC, Raymond J, Roy D. Association between posterior fossa arteriovenous malformations and prenidal aneurysm rupture: potential impact on management. Neurosurg Focus . 2014; 37( 3): E4. Google Scholar CrossRef Search ADS PubMed  29. Orning J, Amin-Hanjani S, Hamade Y et al.   Increased prevalence and rupture status of feeder vessel aneurysms in posterior fossa arteriovenous malformations. J Neurointerv Surg . 2016; 8( 10): 1021- 1024. Google Scholar CrossRef Search ADS PubMed  30. Kaptain GJ, Lanzino G, Do HM, Kassell NF. Posterior inferior cerebellar artery aneurysms associated with posterior fossa arteriovenous malformation: report of five cases and literature review. Surg Neurol . 1999; 51( 2): 146- 152. Google Scholar CrossRef Search ADS PubMed  31. Al-Jehani H, Tampieri D, Cortes M, Melançon D. Re-growth of a posterior inferior cerebellar artery aneurysm after resection of the associated posterior fossa arteriovenous malformation. Interv Neuroradiol . 2014; 20( 1): 61- 66. Google Scholar CrossRef Search ADS PubMed  32. RANKIN J. Cerebral vascular accidents in patients over the age of 60. II. Prognosis. Scott Med J . 1957; 2( 5): 200- 215. Google Scholar CrossRef Search ADS PubMed  33. United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: interim results. UK-TIA Study Group. Br Med J (Clin Res Ed) . 1988; 296( 6618): 316- 320. CrossRef Search ADS PubMed  34. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke . 1988; 19( 5): 604- 607. Google Scholar CrossRef Search ADS PubMed  35. Azzam CJ. Growth of multiple peripheral high flow aneurysms of the posterior inferior cerebellar artery associated with a cerebellar arteriovenous malformation. Neurosurgery . 1987; 21( 6): 934- 939. Google Scholar CrossRef Search ADS PubMed  36. Cunha e Sa MJ, Stein BM, Solomon RA, McCormick PC. The treatment of associated intracranial aneurysms and arteriovenous malformations. J Neurosurg . 1992; 77( 6): 853- 859. Google Scholar CrossRef Search ADS PubMed  37. Spetzler RF, Wilson CB, Weinstein P, Mehdorn M, Townsend J, Telles D. Normal perfusion pressure breakthrough theory. Clin Neurosurg . 1978; 25: 651- 672. Google Scholar CrossRef Search ADS PubMed  38. Alexander MD, Connolly ES, Meyers PM. Revisiting normal perfusion pressure breakthrough in light of hemorrhage-induced vasospasm. World J Radiol . 2010; 2( 6): 230- 232. Google Scholar CrossRef Search ADS PubMed  39. Molyneux AJ1, Kerr RS, Yu LM et al.   International subarachnoid aneurysm trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised comparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet . 2005; 366( 9488): 809- 817. Google Scholar CrossRef Search ADS PubMed  Copyright © 2017 by the Congress of Neurological Surgeons http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Operative Neurosurgery Oxford University Press

Risk Assessment of Hemorrhage of Posterior Inferior Cerebellar Artery Aneurysms in Posterior Fossa Arteriovenous Malformations

Loading next page...
 
/lp/ou_press/risk-assessment-of-hemorrhage-of-posterior-inferior-cerebellar-artery-HxK6BJfj9r
Publisher
Oxford University Press
Copyright
Copyright © 2017 by the Congress of Neurological Surgeons
ISSN
2332-4252
eISSN
2332-4260
D.O.I.
10.1093/ons/opx120
Publisher site
See Article on Publisher Site

Abstract

Abstract BACKGROUND Posterior fossa arteriovenous malformations (AVMs) are associated with increased risk of rupture and severe consequences from such rupture. The hemorrhagic risk of prenidal aneurysms (anr) on the posterior inferior cerebellar artery (PICA) may exceed that of the AVM in posterior fossa AVMs fed by PICA (PICA-AVM). OBJECTIVE To characterize the relative risks of aneurysm and AVM hemorrhage in patients with posterior fossa AVMs. METHODS We retrospectively reviewed patients diagnosed with AVM. Patients with posterior fossa AVMs were divided into 3 groups: PICA-AVM with prenidal aneurysm (PICA-AVM-anr group), PICA-AVM without prenidal aneurysm (PICA-AVM group), and AVMs without PICA feeder with/without aneurysm (AVM-only group). Patient and lesion characteristics and treatment outcomes were compared. ANOVA and chi squared tests were used for statistical analyses. RESULTS Our cohort included 85 patients. Mean age was 45.3 ± 18.1 yr, with 43(50.6%) female patients. Fifty-one patients (60.0%) had hemorrhagic presentation, and 27 (31.8%) experienced acute hydrocephalus. Patients in the PICA-AVM-anr group (n = 11) were more likely to present with aneurysmal subarachnoid hemorrhage (SAH; P = .005) and less likely to have AVM rupture (P < .001). Ten (90.9%) patients presented with hemorrhage, 6 (60.0%) of which resulted from aneurysm rupture. Of these 6, 5 (83.3%) had acute hydrocephalus. No patients with AVM rupture had hydrocephalus. Eight (72.7%) received aneurysm treatment prior to AVM treatment. There were no significant differences in post-treatment outcomes dependent on treatment order. CONCLUSION In addition to relatively higher risk of AVM rupture from infratentorial location and prenidal aneurysm, a higher risk of aneurysm rupture rather than AVM rupture was observed in patients with PICA-AVM-anr complex. Arteriovenous malformation (AVM), Feeding artery, Posterior fossa, Posterior inferior cerebellar artery (PICA), Prenidal aneurysm, Feeding ABBREVIATIONS ABBREVIATIONS anr aneurysm AVM arteriovenous malformation CT computed tomography ICH intracerebral hemorrhage IVH intraventricular hemorrhage MRI magnetic resonance imaging mRS modified Rankin score PICA posterior inferior cerebellar artery SAH subarachnoid hemorrhage Brain arteriovenous malformations (AVMs) commonly manifest with hemorrhage as the initial presentation.1-3 Studies on the natural history of AVMs suggest that the risk of hemorrhagic presentation is 30% to 82%, with a subsequent annual risk of 1.9% to 4.6% in untreated lesions.1,2,4-7 The risk is increased with certain AVM characteristics, including smaller size, associated aneurysms, deep venous drainage, and location.8-10 Posterior fossa AVMs, specifically, have been shown to be more prone to rupture than supratentorial AVMs.11-15 Meta-analysis of patients with posterior fossa AVMs demonstrated 84% risk of hemorrhagic presentation,14 with subsequent annual rupture rate ranging from 7.5% to 11.6% in the first 5 yr.16-18 Therefore, timely diagnosis and treatment of infratentorial lesions are imperative. Surgical treatment risk of posterior fossa AVMs is higher than that of supratentorial AVMs, both due to technical difficulty and postoperative risk. The technical challenge involves the preservation of nearby cranial nerves and brainstem nuclei, while navigating narrow surgical corridors.19-21 Moreover, limited posterior fossa space increases the risk of postsurgical hydrocephalus and poor functional outcomes.22,23 In association with feeding artery aneurysms, the selection of treatment strategy is further complicated by the comparative risk of hemorrhage between the AVM and aneurysm. Feeding artery aneurysms are independent predictors of AVM rupture and subsequent poor outcome, with a 6% attributable risk to hemorrhagic presentation.15,24-27 In some cases, the feeding artery aneurysm is the source of the initial bleed,28-30 and exceedingly high risk of prenidal aneurysmal rupture has been reported in aneurysms on the posterior inferior cerebellar artery (PICA) feeding artery of posterior fossa AVMs.30 In light of these considerations, significant concerns arise regarding treatment priorities between AVMs supplied by PICA (PICA-AVM) and associated prenidal feeding artery aneurysms (PICA-AVM-anr). Better understanding of the risk of hemorrhage and postoperative outcomes of these lesions is critical in making informed decisions. The current literature is limited regarding PICA-AVM combined with feeding artery aneurysms, with only 30 cases reported to date.31 The goal of this study, therefore, is to focus on the subset of patients with this combination, and to compare the hemorrhagic risk and post-treatment outcomes of each. We hypothesize that despite an increased risk of rupture associated with infratentorial location and prenidal aneurysms, the risk of AVM hemorrhage may be overshadowed by the exceeding risk of aneurysmal hemorrhage in patients with PICA-AVM-anr. METHODS Study Population This study is designed as a retrospective cohort study. For this IRB-approved study, we retrospectively reviewed the electronic medical records of 683 patients diagnosed with brain AVMs evaluated at our institution between January 1990 and December 2013. Patient data were collected via IRB-approved database, and consent was exempted given the retrospective nature of this study. Patients with hereditary hemorrhagic telangiectasia, missing baseline information, or lost to follow-up were excluded. The patient selection flowchart is depicted in Figure 1. We only included patients with posterior fossa (infratentorial) AVMs, and the study cohort was divided into 3 groups: PICA-AVM with prenidal aneurysm (PICA-AVM-anr group), PICA-AVM without prenidal aneurysm (PICA-AVM group), and AVMs without PICA feeding artery with/without aneurysm (AVM-only group). Prenidal aneurysm in the PICA-AVM-anr group included both distal feeding aneurysms and those located at the PICA origin. FIGURE 1. View largeDownload slide Flowchart of patient selection process. FIGURE 1. View largeDownload slide Flowchart of patient selection process. Definition of Variable and Outcome Demographic information included sex, age, and race. Age was defined as age at the time of AVM diagnosis. Type of presenting hemorrhage was determined via direct image review or documentation, with AVM rupture defined by intracerebral hemorrhage (ICH), subarachnoid hemorrhage (SAH), and intraventricular hemorrhage (IVH). The source of hemorrhage, if present, was specified as AVM or aneurysm based on image interpretations by both attending neurosurgeon and radiologist. In general, AVM rupture usually presents with ICHs that are largely intraparenchymal and correspond with the location of the AVM. In contrast, aneurysmal hemorrhage predominantly presents with SAH with or without IVH and correspond with the location of the aneurysm. A representative case is presented in Figure 2 to depict the computed tomography (CT) appearance of hemorrhagic presentation with AVM as the primary source of bleed. This case can be contrasted with Figure 3 (PICA-AVM-anr with aneurysm as primary source of bleeding) to distinguish the different appearance on CT. Presence of acute hydrocephalus upon presentation and baseline symptoms (headaches, nausea/vomiting, and dizziness) were noted. AVM lesion characteristics, including nidus size, deep venous drainage, and location, were assessed via baseline angiograms, CT images, and magnetic resonance imaging (MRI). Spetzler–Martin grading was then determined using these 3 parameters. Other angiographic features, such as presence of intranidal aneurysms or other feeding arteries, were also included in the description of each AVM. Morphology and size for each aneurysm were described. FIGURE 2. View largeDownload slide In order to illustrate the difference of hemorrhagic pattern between aneurysmal rupture and AVM rupture, we provided an additional case of hemorrhagic presentation with ruptured AVM in a patient with PICA-AVM (without prenidal aneurysm) to contrast with the CT appearance of Figure 3. A, Axial CT showing predominant left cerebellar intraparenchymal hemorrhage with heterogeneous high-density ICH. B, ICH breaking into the fourth ventricle forming IVH. C, Although with SAH and IVH, which is similar to Figure 3, the evolution of ICH to ventricular breakthrough forming into IVH/SAH is representative of AVM rupture, which is distinctly different from an aneurysmal bleeding demonstrated in Figure 3. D, Left vertebral injection showing a large left AVM with SCA and PICA feeders. FIGURE 2. View largeDownload slide In order to illustrate the difference of hemorrhagic pattern between aneurysmal rupture and AVM rupture, we provided an additional case of hemorrhagic presentation with ruptured AVM in a patient with PICA-AVM (without prenidal aneurysm) to contrast with the CT appearance of Figure 3. A, Axial CT showing predominant left cerebellar intraparenchymal hemorrhage with heterogeneous high-density ICH. B, ICH breaking into the fourth ventricle forming IVH. C, Although with SAH and IVH, which is similar to Figure 3, the evolution of ICH to ventricular breakthrough forming into IVH/SAH is representative of AVM rupture, which is distinctly different from an aneurysmal bleeding demonstrated in Figure 3. D, Left vertebral injection showing a large left AVM with SCA and PICA feeders. FIGURE 3. View largeDownload slide A, Preoperative axial CT showing hemorrhage in the fourth ventricle. B, Preoperative CT showing IVH in the lateral ventricles. C, Anterior-posterior (AP) view of a left vertebral injection on digital subtraction angiography (DSA) demonstrating both the prenidal aneurysm (solid black arrow) and the AVM nidus (dashed black arrow). D, Lateral view of the aneurysm and AVM. E, AP view on DSA showing obliteration of both the aneurysm and AVM following aneurysm embolization and AVM resection. F, Postoperative lateral view DSA. FIGURE 3. View largeDownload slide A, Preoperative axial CT showing hemorrhage in the fourth ventricle. B, Preoperative CT showing IVH in the lateral ventricles. C, Anterior-posterior (AP) view of a left vertebral injection on digital subtraction angiography (DSA) demonstrating both the prenidal aneurysm (solid black arrow) and the AVM nidus (dashed black arrow). D, Lateral view of the aneurysm and AVM. E, AP view on DSA showing obliteration of both the aneurysm and AVM following aneurysm embolization and AVM resection. F, Postoperative lateral view DSA. Functional status at baseline and at last follow-up was assessed using modified Rankin scale (mRS).32-34 Follow-up duration was defined as the interval between AVM diagnosis and last follow-up. AVM obliteration was determined using angiogram, CT image, or MRI at last follow-up. Symptoms (headaches, nausea/vomiting, and dizziness) at last follow-up were determined. Statistical Analysis Patient demographic and angiographic characteristics were compared across different groups. ANOVA and χ2 test were used to analyze continuous and categorical variables respectively. Data analysis was performed using R Statistical Software (Version 3.1.1, 2013, Vienna, Austria). Statistical significance was defined as P < .05. All P values were reported as 2-sided. RESULTS Study Cohort Characteristics A total of 93 patients with posterior fossa AVMs were retrieved from 683 patients. After applying our inclusion and exclusion criteria, our cohort consisted of 85 patients. Mean age of all patients was 45.3 ± 18.1 yr, with 43 (50.6%) female patients. Fifty-one patients (60.0%) had hemorrhagic presentation, and 27 (31.8%) patients experienced acute hydrocephalus. Among the other 34 patients without hemorrhagic presentation, 5 were incidentally diagnosed without any presenting symptoms, and the rest of the patients (n = 29) presented with nonhemorrhagic symptoms such as seizures, persistent headaches, visual disturbances, or cerebellar symptoms that prompted the patient for a neurological workup leading to the discovery of the AVM. Within our study cohort, there were 11 (12.9%) patients in the PICA-AVM-anr group, 32 (37.6%) patients in the PICA-AVM group, and 42 (49.4%) patients in the AVM-only group. Age, gender, and race were similarly distributed across the 3 groups (P = .603, .453, and .287, respectively). There was no significant difference in the proportion of patients presenting with ICH among the 3 groups (P = .191). However, the PICA-AVM and AVM-only groups had significantly fewer patients with presenting SAH than the PICA-AVM-anr group (P = .005). The PICA-AVM and AVM-only groups also showed a trend of fewer IVH than the PICA-AVM-anr group (P = .061). In the PICA-AVM-anr group, 36.4% of patients presented with AVM rupture, compared to 43.8% in the PICA-AVM group and 57.1% in the AVM-only group. For other patients without AVM rupture at presentation, only 1 patient (9.1%) in the PICA-AVM-anr group had nonhemorrhagic presentation, whereas the majority presented with aneurysmal rupture (n = 6, 54.5%). This is in contrast with a 56.3% and 33.3% nonhemorrhagic presentation in the PICA-AVM and AVM-only groups, and only 7.1% aneurysmal rupture in group 3 (P < .001). Despite a higher risk of hemorrhagic presentation in the PICA-AVM-anr group, patients in all 3 groups were equally likely to develop acute hydrocephalus (P = .850). Patients in the PICA-AVM-anr group were more likely to receive surgery ± embolization of the AVM, whereas the patients in the PICA-AVM and AVM-only groups were more likely to receive radiation ± embolization of the AVM. More patients in the PICA-AVM-anr group received embolization as the only treatment modality in regard to the AVM compared to patients in the other 2 groups (P = .006). A description of the comparison is presented in Table 1. TABLE 1. Comparison of Patient and Lesion Characteristics and Treatment Selection Between Different Groups   Total  PICA-AVM-anr  PICA-AVM  AVM-only    Parameters  (n = 85)  (n = 11)  (n = 32)  (n = 42)  P value  Age, mean (sd)  45.25 (18.15)  55.94 (6.89)  39.84 (16.24)  46.58 (19.98)  .603  Gender, male, n (%)  42 (49.4)  6 (54.5)  13 (40.6)  23 (54.8)  .453  Race          .287   White, n (%)  53 (62.4)  5 (45.5)  19 (59.4)  29 (69.0)     Black, n (%)  26 (30.6)  6 (54.5)  11 (34.4)  9 (21.4)     Other, n (%)  6 (7.1)  0 (0.0)  2 (6.3)  4 (9.5)    Spetzler–Martin grades, n (%)          .031*   Grade 1  17 (20.0)  5 (45.5)  6 (18.8)  6 (14.3)     Grade 2  31 (36.5)  4 (36.4)  16 (50.0)  11 (26.2)     Grade 3  30 (35.3)  1 (9.1)  9 (28.1)  20 (47.6)     Grade 4  5 (5.9)  0 (0.0)  1 (3.1)  4 (9.5)     Grade 5  2 (2.4)  1 (9.1)  0 (0.0)  1 (2.4)    AVM max size (cm), mean (sd)  2.34 (1.45)  2.31 (1.39)  2.75 (1.17)  2.03 (1.57)  .185   ICH  40 (47.1)  6 (54.5)  11 (34.4)  23 (54.8)  .191   SAH  15 (17.6)  6 (54.5)  5 (15.6)  4 (9.5)  .005   IVH  27 (31.8)  7 (63.6)  9 (28.1)  11 (26.2)  .061  AVM or aneurysmal rupture          <.001   Aneurysm, n (%)  9 (10.6)  6 (54.5)  0 (0.0)  3 (7.1)     AVM, n (%)  42 (49.4)  4 (36.4)  14 (43.8)  24 (57.1)     None, n (%)  34 (40.0)  1 (9.1)  18 (56.3)  15 (35.7)    Acute hydrocephalus  27 (31.8)  4 (36.4)  9 (28.1)  14 (33.3)  .850  Pretreatment mRS, mean (sd)  1.60 (1.10)  1.45 (0.99)  1.69 (1.21)  1.57 (1.02)  .956  PICA aneurysm  11 (12.9)  11 (100.0)  0 (0.0)  0 (0.0)  <.001  PICA ANR treatment before AVM  8 (9.4)  8 (72.7)  0 (0.0)  0 (0.0)  <.001  AVM treatment, n (%)          .006   Surgery ± embolization  14 (16.5)  4 (36.4)  7 (21.9)  3 (7.1)     Radiation ± embolization  39 (45.9)  1 (9.1)  16 (50.0)  22 (52.4)     Surgery + radiation  2 (2.4)  0 (0.0)  0 (0.0)  2 (4.8)     Embolization  7 (8.2)  3 (27.3)  3 (9.4)  1 (2.4)     Observation  23 (27.1)  3 (27.3)  6 (18.8)  14 (33.3)    AVM obliteration          .421   Yes, n (%)  29 (34.1)  6 (54.5)  10 (31.3)  13 (31.0)     No, n (%)  41 (48.2)  5 (45.5)  16 (50.0)  20 (47.6)     Unknown, n (%)  15 (17.6)  0 (0.0)  6 (18.8)  9 (21.4)      Total  PICA-AVM-anr  PICA-AVM  AVM-only    Parameters  (n = 85)  (n = 11)  (n = 32)  (n = 42)  P value  Age, mean (sd)  45.25 (18.15)  55.94 (6.89)  39.84 (16.24)  46.58 (19.98)  .603  Gender, male, n (%)  42 (49.4)  6 (54.5)  13 (40.6)  23 (54.8)  .453  Race          .287   White, n (%)  53 (62.4)  5 (45.5)  19 (59.4)  29 (69.0)     Black, n (%)  26 (30.6)  6 (54.5)  11 (34.4)  9 (21.4)     Other, n (%)  6 (7.1)  0 (0.0)  2 (6.3)  4 (9.5)    Spetzler–Martin grades, n (%)          .031*   Grade 1  17 (20.0)  5 (45.5)  6 (18.8)  6 (14.3)     Grade 2  31 (36.5)  4 (36.4)  16 (50.0)  11 (26.2)     Grade 3  30 (35.3)  1 (9.1)  9 (28.1)  20 (47.6)     Grade 4  5 (5.9)  0 (0.0)  1 (3.1)  4 (9.5)     Grade 5  2 (2.4)  1 (9.1)  0 (0.0)  1 (2.4)    AVM max size (cm), mean (sd)  2.34 (1.45)  2.31 (1.39)  2.75 (1.17)  2.03 (1.57)  .185   ICH  40 (47.1)  6 (54.5)  11 (34.4)  23 (54.8)  .191   SAH  15 (17.6)  6 (54.5)  5 (15.6)  4 (9.5)  .005   IVH  27 (31.8)  7 (63.6)  9 (28.1)  11 (26.2)  .061  AVM or aneurysmal rupture          <.001   Aneurysm, n (%)  9 (10.6)  6 (54.5)  0 (0.0)  3 (7.1)     AVM, n (%)  42 (49.4)  4 (36.4)  14 (43.8)  24 (57.1)     None, n (%)  34 (40.0)  1 (9.1)  18 (56.3)  15 (35.7)    Acute hydrocephalus  27 (31.8)  4 (36.4)  9 (28.1)  14 (33.3)  .850  Pretreatment mRS, mean (sd)  1.60 (1.10)  1.45 (0.99)  1.69 (1.21)  1.57 (1.02)  .956  PICA aneurysm  11 (12.9)  11 (100.0)  0 (0.0)  0 (0.0)  <.001  PICA ANR treatment before AVM  8 (9.4)  8 (72.7)  0 (0.0)  0 (0.0)  <.001  AVM treatment, n (%)          .006   Surgery ± embolization  14 (16.5)  4 (36.4)  7 (21.9)  3 (7.1)     Radiation ± embolization  39 (45.9)  1 (9.1)  16 (50.0)  22 (52.4)     Surgery + radiation  2 (2.4)  0 (0.0)  0 (0.0)  2 (4.8)     Embolization  7 (8.2)  3 (27.3)  3 (9.4)  1 (2.4)     Observation  23 (27.1)  3 (27.3)  6 (18.8)  14 (33.3)    AVM obliteration          .421   Yes, n (%)  29 (34.1)  6 (54.5)  10 (31.3)  13 (31.0)     No, n (%)  41 (48.2)  5 (45.5)  16 (50.0)  20 (47.6)     Unknown, n (%)  15 (17.6)  0 (0.0)  6 (18.8)  9 (21.4)    *Significant variables (P < .050) View Large Subcohort of PICA-AVM With Feeder Aneurysm There were 11 patients with PICA-AVM and prenidal PICA-feeding artery aneurysms, and detailed patient characteristics are summarized in Table 2. Mean age for these patients with PICA-AVM-anr was 55.9 ± 18.5 yr, with 6 (54.5%) male patients. There were 5 patients with Spetzler–Martin grade I AVM (45.5%), 4 patients with grade II (36.4%), and 1 patient each for grades IV and V (9.1%). None of the patients harbored intranidal aneurysms in the AVM. Ten (90.9%) patients had a hemorrhagic presentation, and 6 (60.0%) of them resulted from aneurysm rupture. Of those with aneurysm rupture, 5 (45.5%) presented with acute hydrocephalus, and all patients with acute hydrocephalus had IVH. No patient with AVM rupture presented with hydrocephalus. All patients who presented with ruptured feeder aneurysm had IVH. Only one patient (9.1%) had a saccular aneurysm. Other patients either had fusiform or irregularly shaped aneurysms. Eight patients (72.7%) had the aneurysm treated before the AVM, of which 5 (62.5%) were treated via embolization and 3 (37.5%) by microsurgical clipping. Nine out of the 11 patients (81.8%) received AVM treatment. Three (27.3%) received surgery only, 3 (27.3%) received embolization only, 1 (9.1%) received radiation only, 1 (9.1%) received surgery and embolization, 1 (9.1%) received radiation and embolization, and 2 (18.2%) were observed without treatment. Complete AVM obliteration was achieved in 6 (54.5%) patients. One patient (9.1%) received treatment for neither the aneurysm nor the AVM. The mean follow-up duration was 52.8 ± 63.5 mo. The average mRS for surgical patients compared to nonsurgical patients is 1.4 vs 1.3. TABLE 2. Characteristics of Patients with PICA Arteriovenous Malformation and PICA Aneurysm (PICA-AVM-anr)   Baseline characteristics  Lesion characteristics    Demographics  Presentation  AVM characteristic  Aneurysm characteristics  Treatment  Follow-up  Pt  Age sex  Race  Symptoms  Hemorrhage  Ruptured lesion  Hydro  Size(mm)/location  DV  Grade  Intranidal anr  Shape  Size (mm)  First  anr  AVM  Length (month)  AVM oblit  Hemorrhage  Symptoms  1  69F  W  H/IB/D/N/BV  +  anr  –  11/right CH  +  II  –  –  2 × 7  AVM  –  S  23  +  _  G  2  49F  W  H/S  +  anr  –  10/left CH  +  II  –  –  6.5  anr  E  E  6  –  –  H/G/HR  3  56M  AA  HR/D/H/IB/BV  –  –  –  29/V  +  II  –  –  –  –  –  –  3  –  –  HR/D/H/BV/G  4  52F  AA  H/D/N/V/IB  +  anr  +  20/V  –  I  –  wide neck  7.8  anr  E  E  2  –  ±  N/V  5  58M  W  H/D  +  AVM  –  15/right CH  –  I  –  +  2.5  anr  C  –  46  +  –  Sensory                        –  4.5                6  58M  AA  H/IB/W  +  AVM  –  34/right CH  +  III  –  –  7 × 8  AVM  –  ER  28  –  –  –  7  43M  AA  H/N/V/IC  +  anr  +  60/Left CH  +  V  –  Irregular  9  anr  E  E  91  –  +  H/W  8  60F  AA  H  +  AVM  –  15/V  +  II  –  –  –  anr  C  R  24  +  –  H/G/sensory  9  60F  AA  H/N/V/IB  +  anr  +  12.5/right CH  –  I  –  –  3  anr  E  ES  35  +  –  –  10  61M  W  H  +  anr  +  28/left CH  –  I  –  –  3.5  anr  E  S  26  +  –  dysphagia  11  48M  W  H/N  +  AVM  –  20/right CH  –  I  –  –  –  anr  C  S  144  +  –  G/BV    Baseline characteristics  Lesion characteristics    Demographics  Presentation  AVM characteristic  Aneurysm characteristics  Treatment  Follow-up  Pt  Age sex  Race  Symptoms  Hemorrhage  Ruptured lesion  Hydro  Size(mm)/location  DV  Grade  Intranidal anr  Shape  Size (mm)  First  anr  AVM  Length (month)  AVM oblit  Hemorrhage  Symptoms  1  69F  W  H/IB/D/N/BV  +  anr  –  11/right CH  +  II  –  –  2 × 7  AVM  –  S  23  +  _  G  2  49F  W  H/S  +  anr  –  10/left CH  +  II  –  –  6.5  anr  E  E  6  –  –  H/G/HR  3  56M  AA  HR/D/H/IB/BV  –  –  –  29/V  +  II  –  –  –  –  –  –  3  –  –  HR/D/H/BV/G  4  52F  AA  H/D/N/V/IB  +  anr  +  20/V  –  I  –  wide neck  7.8  anr  E  E  2  –  ±  N/V  5  58M  W  H/D  +  AVM  –  15/right CH  –  I  –  +  2.5  anr  C  –  46  +  –  Sensory                        –  4.5                6  58M  AA  H/IB/W  +  AVM  –  34/right CH  +  III  –  –  7 × 8  AVM  –  ER  28  –  –  –  7  43M  AA  H/N/V/IC  +  anr  +  60/Left CH  +  V  –  Irregular  9  anr  E  E  91  –  +  H/W  8  60F  AA  H  +  AVM  –  15/V  +  II  –  –  –  anr  C  R  24  +  –  H/G/sensory  9  60F  AA  H/N/V/IB  +  anr  +  12.5/right CH  –  I  –  –  3  anr  E  ES  35  +  –  –  10  61M  W  H  +  anr  +  28/left CH  –  I  –  –  3.5  anr  E  S  26  +  –  dysphagia  11  48M  W  H/N  +  AVM  –  20/right CH  –  I  –  –  –  anr  C  S  144  +  –  G/BV  Age/sex: M: Male, F: Female; Race: W: White, AA: African American; Symptoms: H: Headache, G: gait disturbance/imbalance, D: dizziness, N: nausea, V: vomiting, BV: blurry vision, S: seizure, HR: hearing disturbances, W: weakness, IC: incontinence; Location: CH: cerebellar hemisphere, V: vermis; Treatment: E: embolization, R: radiation, S: surgery, C: microsurgical clipping. View Large All patients presented with headaches. The next most common presenting symptom was imbalance (45.5%), followed by dizziness and nausea (36.4%). Other less common symptoms included blurry vision (18.2%), seizure (9.1%), hearing disturbances (9.1%), weakness (9.1%), and incontinence (9.1%). At last follow-up, the most common symptoms were headaches (36.4%) and imbalance (27.3%). Other symptoms included hearing disturbance, blurry vision, weakness, dysphagia, gait imbalance, and sensory-related symptoms. Case Presentation The patient was a 61-yr-old male who presented with severe headache. Head CT showed SAH with IVH and acute hydrocephalus (Figures 3A and 3B). Angiography revealed a left cerebellar AVM with the nidus measuring 2.8 × 1.0 × 1.0 cm, supplied by multiple distal branches of the left PICA and SCA (Figures 3C and 3D). Venous drainage was superficial via a left tentorial vein to the left transverse sinus. Source of hemorrhage was determined as a ruptured flow-related aneurysm arising from a distal branch of the left PICA. Embolization on the same day targeted the 2 feeding artery branches from the left PICA to the AVM including the parent branch of the PICA aneurysm, resulting in a 30% to 40% nidal volume reduction. Repeat angiography after recovery at 3-mo showed a 2.1 × 1.1 cm AVM in the left cerebellar hemisphere. AVM resection via a suboccipital approach was performed with intraoperative angiography for confirmation AVM obliteration. At 3-mo follow-up, the patient was almost back to baseline, with some residual dysphagia. Repeat angiography at 1 yr showed no residual or recurrent AVM or aneurysm (Figures 3E and 3F). DISCUSSION AVMs with associated feeding artery aneurysms are at increased risk of hemorrhage, although the relative risk of rupture of the AVM compared to the prenidal aneurysm is unclear. The elucidation of the comparative risk between these 2 lesions is critical for treatment decision making, as the core aim of an aneurysm-first or AVM-first treatment approach is to prevent subsequent hemorrhage during the interval between treatments. This study aims to determine this risk with a specific focus on PICA-AVM-anr complex, with a hypothesis derived from clinical impression that more cases of aneurysm rupture were seen compared to AVM rupture. Our study included 11 patients with PICA-AVM-anr complex among a total of 85 patients with posterior fossa AVMs, representing one of the largest series reported to date.30,31,35 Confirming our hypothesis, the majority of patients with both PICA feeding AVMs and PICA aneurysms (PICA-AVM-anr) presented with hemorrhage (90.9%), of which 60% were secondary to the aneurysm rupture. This is in concordance with a literature review of 27 PICA-AVM-anr cases by Kaptain and authors,30 in which 26 patients presented with hemorrhage, and 61.5% of all hemorrhaged cases were confirmed to present with aneurysmal bleeding, with only 7.7% of the hemorrhages attributable to AVM. Based on our results and previous reports that the risk of aneurysm rupture may overshadow AVM rupture, this unique disease combination may distinguish itself from other posterior fossa AVMs or AVM-anr complexes supplied by other feeding arteries, and deserves a thorough re-evaluation of best therapeutic approach to minimize overall risk of hemorrhage. There is currently no consensus on whether the AVM should be treated first or the associated feeding artery aneurysm. Demonstration of aneurysm recession following AVM resection in some reported cases, in particular cases with smaller prenidal aneurysms, suggests that AVMs should be treated before associated prenidal aneurysms in this cohort of patients. This can be explained by 2 theories: restoration of normal blood flow, which reduces aneurysmal dilatation, or flow reduction that promotes aneurysm sac thrombosis.35,36 However, the reduction in blood flow in feeder vessels following AVM resection may also increase the risk of subsequent hemorrhage via the normal perfusion pressure breakthrough theory, especially in larger or high-grade (grade 4 or 5) AVMs where blood-steal phenomenon is predominant. Normal perfusion pressure breakthrough theory suggests that the presence of AVM results in the chronic reduction in blood flow to normal tissue. As a result, the vessels supplying them dilate and lose their ability to autoregulate. Following AVM excision, the transient increase in perfusion to these vessels, coupled with their inability to vasoconstrict, predisposes the patient to further hemorrhage from the untreated feeder aneurysm.37,38 Therefore, for selected patients identified with high risk of aneurysm rupture, the aneurysm treatment should come before AVM resection. Existing studies on posterior fossa AVMs with feeder aneurysms also support this recommendation.28,29 It is unknown why PICA-AVM-anr bear a significantly higher risk of aneurysm rupture than other AVM-anr complexes; 1 possibility is that PICA-feeding artery aneurysms alone may have an overall higher risk of hemorrhage compared to other posterior circulation aneurysms regardless of existence of AVMs. From the results of the International Subarachnoid Aneurysm Trial, PICA feeding artery aneurysms accounted for 53.4% of all posterior fossa aneurysm ruptures. Interestingly, the second highest risk of ruptured presentation was basilar apex aneurysms (29.3%), which is with minimal association with feeding of AVM. In contrast, for superior cerebellar arteries or posterior cerebral arteries that have the potential to feed AVMs, only 9 cases (0.4%) were noted to have ruptured in 2143 cases.39 It is therefore reasonable to assume that the higher risk of aneurysmal hemorrhage in PICA-AVM-anr is likely to be related intrinsically to PICA itself, either due to particular anatomic reason or hemodynamic characteristics, as opposed to the impact from the combination of PICA-AVM. Future studies are warranted to further elucidate the underlying cause of this phenomenon. In our study, 8 out of 10 patients (80%) who underwent treatment had received aneurysm treatment prior to AVM treatment. Two patients had recurrent hemorrhage after aneurysm treatment. The first patient (patient 4; Table 2) was a 52-yr-old female with a grade I AVM and a 7.8 mm PICA aneurysm that received coil embolization. She experienced mild acute intraparenchymal hemorrhage 4 d after aneurysm treatment. The second patient (patient 7; Table 2) was a 43-yr-old male with a Spetzler–Martin grade V AVM and a 9 mm aneurysm that was embolized. He had recurrent hemorrhage 3 yr after initial aneurysm embolization. Angiography at the time showed persistent AVM without significant feeding artery aneurysms as the source of bleed. The patient continued to experience weakness and headaches until last follow-up. These results suggest that despite aneurysm treatment, there is still risk for recurrent hemorrhage. In 1 patient (patient 5; Table 2), the AVM spontaneously thrombosed on angiography at 6 mo after aneurysm treatment. This supports the theory that treating feeder aneurysms prior to treating AVM may reduce the flow to the associated AVM. This may result in a decrease in AVM size or, as in this case, lead to thrombosis. However, it is important to note that spontaneous thrombosis following embolization of a feeder aneurysm should not be the expectation following aneurysm treatment. Definitive treatment of the AVM should still be sought. Of the 2 patients who received AVM treatment first (patients 1 and 6; Table 2), 1 was asymptomatic at 6-mo follow-up, while the other experienced elevated intracranial pressure and residual imbalance. These findings were not significantly better or worse than follow-up symptoms of those who received aneurysm treatment first. This suggests that there is no obvious difference in functional outcome between those with aneurysm treated first and those with AVM treated first. Within this population, acute hydrocephalus arising after hemorrhage is potentially catastrophic due to proximity of the fourth ventricle and cerebral aqueduct. Therefore, correlation of hydrocephalus risk with AVM and aneurysm rupture is essential in formulating treatment plans. Our cohort included 4 patients (36.4%) who presented with acute hydrocephalus, all of whom were found to have IVH. All 4 patients hemorrhaged because of PICA feeding artery aneurysm rupture. No patients who presented with AVM rupture experienced acute hydrocephalus. Furthermore, all patients with aneurysm rupture were found to have IVH, including those who did not meet criteria for acute hydrocephalus. This suggests that PICA aneurysm rupture may be correlated with a higher complication rate of hydrocephalus than PICA-AVM rupture—a likely consequence of development of IVH, which predisposes patients to acute hydrocephalus. There was no significant difference in residual symptoms at last follow-up. Limitations There are several study limitations that must be highlighted. As a single-institution study, sampling bias limits the generalizability of our results. Furthermore, there is also attrition bias due to the exclusion of patients who were lost to follow-up or had missing critical information. The small sample size makes it difficult to reach sufficient statistical power. Another potential limitation is the variability of treatment modality demonstrated in this study cohort. As treatment selection is determined by a shared decision-making process with patient preference having a significant weight in our institution, a variability of treatment was observed, especially for low-grade patients where surgical resection was regarded as the best treatment. However, this also provided us with the advantage of comparing disease characteristics by different treatment modalities with lower risk of selection bias. CONCLUSION In this report of a rare combination of PICA-AVM-anr, patients are more likely to present with hemorrhage from the associated PICA feeding artery aneurysm as opposed to the posterior fossa AVM. The concordance of our results with previous studies warrants a thorough evaluation of management strategy when addressing this disease combination. We suggest that a PICA-AVM-anr complex should be considered for an aneurysm-first strategy to reduce risk of aneurysm rupture. AVM treatment should follow promptly after aneurysm treatment in order to reduce the risk of hemorrhage from the AVM that is potentially attributable to disturbance of hemodynamics after prenidal aneurysm treatment. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Brown RD, Wiebers DO, Forbes G et al.   The natural history of unruptured intracranial arteriovenous malformations. J Neurosurg . 1988; 68( 3): 352- 357. Google Scholar CrossRef Search ADS PubMed  2. Fleetwood IG, Steinberg GK. Arteriovenous malformations. Lancet . 2002; 359( 9309): 863- 873. Google Scholar CrossRef Search ADS PubMed  3. Ondra SL, Troupp H, George ED, Schwab K. The natural history of symptomatic arteriovenous malformations of the brain: a 24-year follow-up assessment. J Neurosurg . 1990; 73( 3): 387- 391. Google Scholar CrossRef Search ADS PubMed  4. da Costa L, Wallace MC, Ter Brugge KG, O’Kelly C, Willinsky RA, Tymianski M. The natural history and predictive features of hemorrhage from brain arteriovenous malformations. Stroke . 2009; 40( 1): 100- 105. Google Scholar CrossRef Search ADS PubMed  5. Graf CJ, Perret GE, Torner JC. Bleeding from cerebral arteriovenous malformations as part of their natural history. J Neurosurg . 1983; 58( 3): 331- 337. Google Scholar CrossRef Search ADS PubMed  6. Gross BA, Du R. Natural history of cerebral arteriovenous malformations: a meta-analysis. J Neurosurg . 2013; 118( 2): 437- 443. Google Scholar CrossRef Search ADS PubMed  7. Hernesniemi JA, Dashti R, Juvela S, Väärt K, Niemelä M, Laakso A. Natural history of brain arteriovenous malformations: a long-term follow-up study of risk of hemorrhage in 238 patients. Neurosurgery . 2008; 63( 5): 829- 831. Google Scholar CrossRef Search ADS   8. Dashe JF. Predictors of hemorrhage in patients with untreated brain arteriovenous malformation. Neurology . 2007; 68( 7): 535; author reply 535. Google Scholar CrossRef Search ADS PubMed  9. Langer DJ, Lasner TM, Hurst RW, Flamm ES, Zager EL, King JT. Hypertension, small size, and deep venous drainage are associated with risk of hemorrhagic presentation of cerebral arteriovenous malformations. Neurosurgery . 1998; 42( 3): 481- 489. Google Scholar CrossRef Search ADS PubMed  10. Stefani MA, Porter PJ, terBrugge KG, Montanera W, Willinsky RA, Wallace MC. Angioarchitectural factors present in brain arteriovenous malformations associated with hemorrhagic presentation. Stroke . 2002; 33( 4): 920- 924. Google Scholar CrossRef Search ADS PubMed  11. Batjer H, Samson D. Arteriovenous malformations of the posterior fossa. Clinical presentation, diagnostic evaluation, and surgical treatment. J Neurosurg . 1986; 64( 6): 849- 856. Google Scholar CrossRef Search ADS PubMed  12. Brown RD, Wiebers DO, Torner JC, O’Fallon WM. Frequency of intracranial hemorrhage as a presenting symptom and subtype analysis: a population-based study of intracranial vascular malformations in Olmsted Country, Minnesota. J Neurosurg . 1996; 85( 1): 29- 32. Google Scholar CrossRef Search ADS PubMed  13. Drake CG, Friedman AH, Peerless SJ. Posterior fossa arteriovenous malformations. J Neurosurg . 1986; 64( 1): 1- 10. Google Scholar CrossRef Search ADS PubMed  14. Arnaout OM, Gross BA, Eddleman CS, Bendok BR, Getch CC, Batjer HH. Posterior fossa arteriovenous malformations. Neurosurg Focus . 2009; 26( 5): E12. Google Scholar CrossRef Search ADS PubMed  15. Khaw AV, Mohr JP, Sciacca RR et al.   Association of infratentorial brain arteriovenous malformations with hemorrhage at initial presentation. Stroke . 2004; 35( 3): 660- 663. Google Scholar CrossRef Search ADS PubMed  16. Kelly ME, Guzman R, Sinclair J et al.   Multimodality treatment of posterior fossa arteriovenous malformations. J Neurosurg . 2008; 108( 6): 1152- 1161. Google Scholar CrossRef Search ADS PubMed  17. Mine S, Hirai S, Ono J, Yamaura A. Risk factors for poor outcome of untreated arteriovenous malformation. J Clin Neurosci . 2000; 7( 6): 503- 506. Google Scholar CrossRef Search ADS PubMed  18. Yamada S, Takagi Y, Nozaki K, Kikuta K, Hashimoto N. Risk factors for subsequent hemorrhage in patients with cerebral arteriovenous malformations. J Neurosurg . 2007; 107( 5): 965- 972. Google Scholar CrossRef Search ADS PubMed  19. Drake CG. Cerebral arteriovenous malformations: considerations for and experience with surgical treatment in 166 cases. Clin Neurosurg . 1979; 26: 145- 208. Google Scholar CrossRef Search ADS PubMed  20. Neacsu A, Ciurea AV. General considerations on posterior fossa arteriovenous malformations (clinics, imaging and therapy). Actual concepts and literature review. J Med Life . 2010; 3( 1): 26- 35. Google Scholar PubMed  21. Solomon RA, Stein BM. Management of arteriovenous malformations of the brain stem. J Neurosurg . 1986; 64( 6): 857- 864. Google Scholar CrossRef Search ADS PubMed  22. Torné R, Rodríguez-Hernández A, Arikan F et al.   Posterior fossa arteriovenous malformations: Significance of higher incidence of bleeding and hydrocephalus. Clin Neurol Neurosurg . 2015; 134: 37- 43. Google Scholar CrossRef Search ADS PubMed  23. Yang W, Wang JY, Caplan JM et al.   Predictors of functional outcome following treatment of posterior fossa arteriovenous malformations. J Clin Neurosci . 2015; 22( 2): 357- 362. Google Scholar CrossRef Search ADS PubMed  24. da Costa L, Thines L, Dehdashti AR et al.   Management and clinical outcome of posterior fossa arteriovenous malformations: report on a single-centre 15-year experience. J Neurol Neurosurg Psychiatry . 2009; 80( 4): 376- 379. Google Scholar CrossRef Search ADS PubMed  25. Abla AA, Nelson J, Rutledge WC, Young WL, Kim H, Lawton MT. The natural history of AVM hemorrhage in the posterior fossa: comparison of hematoma volumes and neurological outcomes in patients with ruptured infra- and supratentorial AVMs. Neurosurg Focus . 2014; 37( 3): E6. Google Scholar CrossRef Search ADS PubMed  26. Stapf C, Mohr JP, Pile-Spellman J et al.   Concurrent arterial aneurysms in brain arteriovenous malformations with haemorrhagic presentation. J Neurol Neurosurg Psychiatry . 2002; 73( 3): 294- 298. Google Scholar CrossRef Search ADS PubMed  27. Thompson RC, Steinberg GK, Levy RP, Marks MP. The management of patients with arteriovenous malformations and associated intracranial aneurysms. Neurosurgery . 1998; 43( 2): 202- 211. Google Scholar CrossRef Search ADS PubMed  28. Kouznetsov E, Weill A, Ghostine JS, Gentric JC, Raymond J, Roy D. Association between posterior fossa arteriovenous malformations and prenidal aneurysm rupture: potential impact on management. Neurosurg Focus . 2014; 37( 3): E4. Google Scholar CrossRef Search ADS PubMed  29. Orning J, Amin-Hanjani S, Hamade Y et al.   Increased prevalence and rupture status of feeder vessel aneurysms in posterior fossa arteriovenous malformations. J Neurointerv Surg . 2016; 8( 10): 1021- 1024. Google Scholar CrossRef Search ADS PubMed  30. Kaptain GJ, Lanzino G, Do HM, Kassell NF. Posterior inferior cerebellar artery aneurysms associated with posterior fossa arteriovenous malformation: report of five cases and literature review. Surg Neurol . 1999; 51( 2): 146- 152. Google Scholar CrossRef Search ADS PubMed  31. Al-Jehani H, Tampieri D, Cortes M, Melançon D. Re-growth of a posterior inferior cerebellar artery aneurysm after resection of the associated posterior fossa arteriovenous malformation. Interv Neuroradiol . 2014; 20( 1): 61- 66. Google Scholar CrossRef Search ADS PubMed  32. RANKIN J. Cerebral vascular accidents in patients over the age of 60. II. Prognosis. Scott Med J . 1957; 2( 5): 200- 215. Google Scholar CrossRef Search ADS PubMed  33. United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: interim results. UK-TIA Study Group. Br Med J (Clin Res Ed) . 1988; 296( 6618): 316- 320. CrossRef Search ADS PubMed  34. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke . 1988; 19( 5): 604- 607. Google Scholar CrossRef Search ADS PubMed  35. Azzam CJ. Growth of multiple peripheral high flow aneurysms of the posterior inferior cerebellar artery associated with a cerebellar arteriovenous malformation. Neurosurgery . 1987; 21( 6): 934- 939. Google Scholar CrossRef Search ADS PubMed  36. Cunha e Sa MJ, Stein BM, Solomon RA, McCormick PC. The treatment of associated intracranial aneurysms and arteriovenous malformations. J Neurosurg . 1992; 77( 6): 853- 859. Google Scholar CrossRef Search ADS PubMed  37. Spetzler RF, Wilson CB, Weinstein P, Mehdorn M, Townsend J, Telles D. Normal perfusion pressure breakthrough theory. Clin Neurosurg . 1978; 25: 651- 672. Google Scholar CrossRef Search ADS PubMed  38. Alexander MD, Connolly ES, Meyers PM. Revisiting normal perfusion pressure breakthrough in light of hemorrhage-induced vasospasm. World J Radiol . 2010; 2( 6): 230- 232. Google Scholar CrossRef Search ADS PubMed  39. Molyneux AJ1, Kerr RS, Yu LM et al.   International subarachnoid aneurysm trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised comparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet . 2005; 366( 9488): 809- 817. Google Scholar CrossRef Search ADS PubMed  Copyright © 2017 by the Congress of Neurological Surgeons

Journal

Operative NeurosurgeryOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off