Non-Aneurysmal Subarachnoid Hemorrhage & Vascular Malformations

Non-Aneurysmal Subarachnoid Hemorrhage & Vascular Malformations

Approximately $15\%$ of spontaneous subarachnoid hemorrhages occur in the absence of an identifiable intracranial aneurysm on initial catheter angiography. Furthermore, intracranial vascular malformations—including arteriovenous malformations ($AVMs$), dural arteriovenous fistulas ($dAVFs$), and cavernous malformations—represent critical neurovascular entities that present with subarachnoid or intraparenchymal hemorrhage, seizures, and progressive focal deficits. Neurocritical care clinicians must distinguish benign non-aneurysmal hemorrhage patterns from high-risk structural vascular malformations that demand targeted surgical or endovascular interventions.

---

Perimesencephalic Non-Aneurysmal SAH (PM-SAH)

Perimesencephalic non-aneurysmal subarachnoid hemorrhage ($PM\text{-}SAH$) accounts for approximately $10\text{--}15\%$ of all spontaneous SAH presentations and up to two-thirds of all angiogram-negative cases.

Diagnostic Criteria (The van Gijn Criteria)

To be classified as true perimesencephalic SAH, the initial non-contrast head CT obtained within 72 hours of symptom onset must fulfill strict topographic anatomical criteria:

1. Epicenter of Hemorrhage: Blood is located immediately anterior to the midbrain or upper pons (interpeduncular, prepontine, or chiasmatic cisterns).
2. Limited Cisternal Extension: Clot may extend symmetrically or asymmetrically into the ambient cisterns, crural cisterns, or basal portions of the Sylvian fissures.
3. Absence of Frank Intraventricular Hemorrhage ($IVH$): No frank intraventricular casting (minor sedimentation of red cells in the occipital horns of the lateral ventricles is permissible).
4. Absence of Lateral Sylvian or Interhemispheric Clot: No blood pooling in the lateral Sylvian fissures and no anterior interhemispheric fissure clot (findings that strongly indicate a ruptured middle cerebral or anterior communicating artery aneurysm).
5. No Intraparenchymal Hematoma: Absence of brain parenchymal clot.

[ Acute Non-Traumatic SAH on NCCT ]
                 │
      ┌──────────┴──────────┐
      ▼                     ▼
[ True Perimesencephalic ] [ Non-Perimesencephalic Pattern ]
• Blood anterior to midbrain • Thick cisternal, lateral Sylvian,
• No frank IVH/ICH             or interhemispheric clot
      │                             │
      ▼                             ▼
[ Initial 3D-CTA / DSA: Neg ] [ Initial 3D-CTA / DSA: Neg ]
      │                             │
      ▼                             ▼
• Venous origin proposed    • High occult aneurysm risk (~10-15%)
• Rebleeding rate ~0%        • Repeat vascular workup considered
• Vasospasm/DCI < 1%           in 7–14 days
• Excellent prognosis        • Full SAH ICU surveillance bundle
• Repeat DSA rarely needed

Etiology, Angiographic Workup & Prognosis

• Etiology: A venous mechanism has been proposed for classic perimesencephalic SAH, but a single mechanism is not established for every case. Confirm that the imaging pattern and vascular evaluation adequately exclude an aneurysm or another cause before using a lower-risk label.

---

Brain Arteriovenous Malformations (AVMs)

Brain arteriovenous malformations are high-flow congenital vascular anomalies comprising direct fistulous connections between feeding cerebral arteries and draining veins through a central dysplastic vascular tangle—the nidus—without an intervening capillary bed.

Clinical Presentation & Hemodynamics

• Hemorrhage ($50\%$): Intraparenchymal hemorrhage ($ICH$), intraventricular hemorrhage ($IVH$), or subarachnoid hemorrhage ($SAH$).
• Seizures ($30\%$): Secondary to cortical gliosis, hemosiderin irritation, or localized ischemia.
• Arterial Steal Syndrome: High-volume, low-resistance arteriovenous shunting diverts blood away from surrounding normal brain parenchyma, precipitating progressive focal neurological deficits.

The Spetzler-Martin Grading Scale

The Spetzler-Martin scale predicts the morbidity and mortality associated with microsurgical resection of brain AVMs:

• Grades I–II (Low Risk): Low surgical morbidity ($< 1\text{--}3\%$); complete microsurgical resection is the definitive standard of care.
• Grade III (Intermediate Risk): Heterogeneous cohort requiring individualized multimodal management (microsurgery, preoperative embolization, or radiosurgery).
• Grades IV–V (High Risk): Significant surgical morbidity and mortality ($20\text{--}40\%$); microsurgical resection is generally contraindicated unless the patient presents with a life-threatening, accessible hematoma.

The ARUBA Trial: Unruptured AVMs

The landmark ARUBA trial (A Randomized Trial of Unruptured Brain Arteriovenous Malformations) compared medical management alone (antiepileptic drugs, blood pressure control) versus interventional therapy (microsurgery, endovascular embolization, stereotactic radiosurgery, or combinations) in adults with unruptured AVMs.

• Primary Endpoint: The trial was halted prematurely after enrolling 226 patients due to overwhelming superiority of medical management alone: the risk of stroke or death was significantly higher in the interventional therapy arm ($30.7\%$) compared to the medical management arm ($10.1\%$), with a hazard ratio of $0.27$ for medical management relative to intervention.
• Long-Term Evidence: Extended 5-year follow-up confirmed persistent disparity. Consequently, guideline-directed neurocritical practice avoids routine prophylactic interventional eradication for unruptured brain AVMs.
• Ruptured AVM Management: In contrast, patients presenting with a ruptured AVM face a rebleeding hazard of $6\text{--}18\%$ in the first year post-bleed (compared to $1\text{--}3\%$/year for unruptured AVMs). Ruptured AVMs require specialist planning for hemorrhage management and possible definitive treatment; timing and modality depend on anatomy, clinical stability, and treatment risk.

---

Dural Arteriovenous Fistulas (dAVFs)

Dural arteriovenous fistulas are acquired, pathological arteriovenous shunts located entirely within the dural leaflets, commonly fed by branches of the external carotid artery (middle meningeal, occipital, ascending pharyngeal arteries) and draining into dural venous sinuses or cortical leptomeningeal veins. They often arise following dural sinus thrombosis, head trauma, or prior craniotomy.

The Borden and Cognard Classifications

---

Cerebral Cavernous Malformations (Cavernomas)

Cerebral cavernous malformations ($CCMs$) are low-flow vascular anomalies consisting of compact clusters of dilated, hyalinized endothelial-lined capillary-like sinusoidal channels ("caverns") devoid of intervening neural parenchyma or mature vessel wall architecture (lacking smooth muscle and internal elastic lamina).

Neuroimaging Hallmarks

• MRI T2-Weighted / FLAIR: Pathognomonic "popcorn" or "mulberry" appearance, with a central core of heterogeneous mixed signal intensity representing hemorrhages of varying ages (thrombosis, methemoglobin, hemosiderin), encapsulated by a complete circumferential rim of profound T2 hypointensity due to hemosiderin deposition.
• Gradient Echo ($GRE$) and Susceptibility-Weighted Imaging ($SWI$): Displays prominent, extensive "blooming" artifact due to magnetic susceptibility from iron breakdown products.
• Catheter Angiography ($DSA$): Angiographically occult. Cavernomas have low, sluggish internal blood flow without arteriovenous shunting and are invisible on catheter digital subtraction angiography.

                    [ Cerebral Cavernous Malformation ]
                                     │
         ┌───────────────────────────┴───────────────────────────┐
         ▼                                                       ▼
[ MRI Pathognomonic Appearance ]                        [ Associated DVA Alert ]
• Central mixed-signal "popcorn" core                   • Found in 30-40% of cases
• Complete hypointense hemosiderin rim                  • Provides functional venous
• Marked blooming on SWI/GRE                              drainage of normal brain
• Angiographically occult on DSA                        • DO NOT RESECT OR OCCLUDE DVA!

Developmental Venous Anomalies (DVAs) & Surgical Principles

In $30\text{--}40\%$ of sporadic cavernomas, a Developmental Venous Anomaly ($DVA$) ("caput medusae" draining into a collector vein) is intimately associated with the lesion.

Natural History and Indications for Intervention

• Hemorrhage Rates:

  • Unruptured incidental cavernomas: Annual bleeding rate of $0.5\text{--}1\%$ per year.
  • Previously ruptured cavernomas: Annual rebleeding rate increases to $4\text{--}5\%$ per year (clustering phenomenon within the first 2–3 years post-hemorrhage).
  • Brainstem cavernomas carry significantly higher neurological morbidity with each bleeding episode due to the high density of cranial nerve nuclei and tracts.

• Surgical Indications: Microsurgical resection is indicated for: (1) surgically accessible cavernomas causing medically refractory epilepsy, (2) progressive focal neurological deficits, or (3) recurrent symptomatic hemorrhages. Stereotactic radiosurgery is reserved as a second-line consideration for deep, surgically inaccessible brainstem lesions with multiple life-threatening hemorrhages.

• Angiographic strategy: Interpret the blood pattern, initial CTA/DSA quality, clinical course, and residual suspicion together. Strictly perimesencephalic bleeding with an adequate negative evaluation has a low subsequent diagnostic yield. Nonperimesencephalic diffuse bleeding has a greater occult vascular-lesion risk and often warrants repeat catheter angiography. The timing and modality are specialist decisions rather than an inflexible universal 7–14-day rule.

• Surveillance: An aneurysmal pattern with a negative first angiogram remains concerning. Thrombosis, vasospasm, technical limitations, or obscuring blood can hide a lesion. Continue appropriate monitoring while completing the diagnostic plan; do not discharge solely because the first angiogram was negative.