RCVS² Score

Why use it

The treatment for RCVS and PACNS is diametrically opposed — immunosuppression (corticosteroids, cyclophosphamide) benefits PACNS but can potentially worsen RCVS by interfering with natural resolution of vasoconstriction. Misdiagnosis carries serious consequences: untreated PACNS can cause progressive strokes and death; unnecessary immunosuppression in RCVS exposes patients to significant adverse effects. The RCVS² score provides an objective, quantitative framework to aid this high-stakes differential diagnosis at presentation, before the 3-month follow-up imaging that confirms RCVS by demonstrating vasoconstriction resolution.

About

The RCVS² Score is a validated clinical decision tool developed by Rocha, Topcuoglu, Silva, and Singhal and published in Stroke in 2019 to help differentiate reversible cerebral vasoconstriction syndrome (RCVS) from other intracranial arteriopathies — most critically from primary angiitis of the central nervous system (PACNS), which requires opposite treatment (immunosuppression vs. avoidance in RCVS). RCVS is characterized by thunderclap headache (TCH) and segmental cerebral vasoconstriction that resolves within 3 months, often triggered by vasoactive substances. PACNS is a granulomatous inflammation of small cerebral vessels requiring high-dose corticosteroids and cyclophosphamide. Distinguishing these two conditions clinically is challenging since both may present with headache, focal deficits, and abnormal cerebral vascular imaging. The RCVS² score integrates five variables to compute a score ranging from −2 to +10: Recurrent thunderclap headache (R), Carotid artery involvement on vascular imaging (C — negative), Vasoconstrictive trigger present (V), female Sex (S), and Subarachnoid hemorrhage (² — negative). At the optimal cut-point of ≥5, the score achieves a sensitivity of 90% and specificity of 99% for RCVS diagnosis. The score was derived in a multicenter cohort of 113 patients (76 RCVS, 37 PACNS) and externally validated.

When to use

• Patients presenting with thunderclap headache and abnormal cerebral vascular imaging (MRA/CTA showing segmental vasoconstriction)
• Differentiating RCVS from PACNS when vascular imaging shows multifocal narrowing of intracranial arteries
• Patients with focal neurological deficits, ischemic stroke, or intracerebral hemorrhage in the context of headache and suspected arteriopathy
• Before initiating immunosuppressive therapy for suspected CNS vasculitis — use RCVS² to confirm low RCVS probability
• NOT a substitute for complete workup including vascular imaging, CSF analysis, and neurology consultation

Formula

Pearls & pitfalls

• The hallmark of RCVS is RECURRENT thunderclap headache (TCH) — multiple separate episodes of TCH are highly characteristic. A single episode of TCH in a low-risk patient may not be RCVS.
• Vasoconstriction in RCVS typically involves MEDIUM and LARGE cerebral arteries (M1, M2, A1, P1, basilar), while PACNS tends to involve SMALL vessels that may not be visible on conventional MRA/CTA — this means a normal conventional angiogram in the context of ongoing neurological deterioration should prompt catheter angiography or brain biopsy to evaluate for small-vessel PACNS.
• Carotid artery involvement is a NEGATIVE predictor for RCVS (−2 points) — RCVS predominantly affects intracranial vessels; carotid involvement is more suggestive of vasculitis extending into extracranial vessels.
• The convexal subarachnoid hemorrhage (SAH) associated with RCVS is typically NON-aneurysmal and located in cortical sulci over the convexities — this is different from aneurysmal SAH which predominates in the basal cisterns. However, convexal SAH also occurs in PACNS, hence it contributes +1 point in the validated RCVS² score.
• Common vasoactive triggers for RCVS: ergotamines, triptans, SSRIs/SNRIs (especially when starting or stopping), sympathomimetics (pseudoephedrine, phenylephrine), illicit stimulants (cocaine, amphetamines, MDMA), cannabis, postpartum state, blood transfusion, and vigorous sexual activity or physical exertion.
• Female sex contributes +1 point because RCVS has a strong female predominance (approximately 70–80% female), particularly in younger women and postpartum cases.
• CSF analysis: RCVS typically shows NORMAL or near-normal CSF (no pleocytosis); PACNS typically shows elevated protein and lymphocytic pleocytosis. A CSF cell count >10 cells/µL should significantly shift the clinical assessment toward PACNS.
• The score was derived against catheter angiography-confirmed diagnosis — MRA and CTA are less sensitive for small vessel involvement and may miss PACNS. When RCVS² is indeterminate (2–4), catheter angiography is often warranted.
• Pregnancy and the postpartum period (up to 6 weeks post-delivery) are the most common triggers for RCVS in young women — the score accounts for this via the female sex item and vasoconstrictive trigger item.

Next steps

RCVS² ≥5 (RCVS very likely): Avoid immunosuppression. Discontinue vasoactive triggers (triptans, SSRIs, SNRIs, decongestants, cannabis, illicit stimulants). Consider calcium channel blockers (nimodipine 60 mg q4h or verapamil) — may reduce headache frequency and severity. Provide analgesics for thunderclap headache. Repeat MRA/CTA at 12 weeks to confirm vasoconstriction resolution (diagnostic confirmation). Avoid strenuous physical activity and Valsalva-provoking activities. RCVS² 2–4 (Indeterminate): Further workup essential. Consider brain and leptomeningeal biopsy to exclude PACNS. CSF analysis (pleocytosis favors PACNS). Multidisciplinary neurology discussion. RCVS² ≤1 (PACNS more likely): Strong consideration of CNS vasculitis workup including brain/meningeal biopsy. CSF analysis for pleocytosis and elevated protein. Consider MRI perfusion imaging. If biopsy confirms granulomatous angiitis: initiate methylprednisolone 1 g IV × 3–5 days followed by prednisone 1 mg/kg/day + cyclophosphamide.

Evidence

Rocha EA et al. RCVS2 score and diagnostic approach for reversible cerebral vasoconstriction syndrome. Neurology. 2019;92(7):e639-e647. PMID:30635484. Derivation cohort: n=113 patients (76 RCVS, 37 PACNS). At cut-point ≥5: sensitivity 90%, specificity 99%, AUC 0.97. External validation in multiple subsequent cohorts confirmed discrimination (AUC 0.92–0.96). Now incorporated into AHA/ASA secondary stroke prevention guidelines and expert consensus for RCVS diagnosis.

References

1. Rocha EA et al. RCVS2 score and diagnostic approach for reversible cerebral vasoconstriction syndrome. Neurology. 2019;92(7):e639-e647. — link
2. Singhal AB et al. Reversible cerebral vasoconstriction syndromes: analysis of 139 cases. Arch Neurol. 2011;68(8):1005-1012. — link
3. Calabrese LH et al. Narrative review: reversible cerebral vasoconstriction syndromes. Ann Intern Med. 2007;146(1):34-44. — link
4. Ducros A. Reversible cerebral vasoconstriction syndrome. Lancet Neurol. 2012;11(10):906-917. — link