14. Adjunctive DCI-Prevention Pharmacotherapy in SAH (Magnesium, Statins, Clazosentan)

Quick Recap

Neurology System — companion to Protocol 3 (Subarachnoid Hemorrhage), which already comprehensively covers nimodipine, the Triple-H revision, and induced-hypertension/angioplasty treatment of established DCI. ⚠️ This protocol is deliberately scoped to cover only what that protocol does not: the substantial body of adjunctive pharmacotherapy tested specifically because nimodipine alone leaves real residual DCI burden — magnesium, statins, and the endothelin receptor antagonist clazosentan. All three have now been tested in large, rigorous phase 3 trials, and the clazosentan story specifically illustrates a clean, instructive dissociation between angiographic/surrogate improvement and patient-centered functional outcome.

1. Definition

Why adjunctive agents beyond nimodipine were pursued: nimodipine is the one agent with a consistent, guideline-endorsed efficacy signal for improving neurologic outcome after aneurysmal SAH (cross-reference Protocol 3, Section 11), but this benefit is incomplete — delayed cerebral ischemia (DCI) remains the dominant driver of secondary disability even with universal nimodipine use, motivating three decades of trials testing other pharmacologic targets specifically aimed at either angiographic vasospasm or the broader DCI process.

The three most extensively tested adjunctive agents, each addressed by a large, definitive phase 3 program:

• Magnesium sulfate: proposed mechanism as a cerebral vasodilator and NMDA-receptor antagonist with neuroprotective potential
• Statins (simvastatin): proposed "pleiotropic" anti-inflammatory and endothelial-protective effects, distinct from cholesterol-lowering — a mechanistic rationale structurally similar to the statins-in-ARDS story addressed elsewhere in this library (cross-reference Statins in Sepsis & ARDS protocol, Respiratory System)
• Clazosentan: a selective endothelin receptor antagonist, targeting endothelin-1's role as a potent vasoconstrictor implicated in angiographic vasospasm specifically

2. Pathophysiology

Endothelin-1 is a potent, sustained vasoconstrictor released after SAH, mechanistically linked to angiographic cerebral vasospasm — clazosentan's proposed mechanism directly targets this specific pathway, and experimental and early-phase clinical data consistently confirmed clazosentan does reduce angiographic vasospasm, a genuinely successful proof-of-mechanism. The critical, recurring question across this entire evidence base is whether reducing the angiographic/surrogate marker (vasospasm severity on imaging) reliably translates into improved patient-centered functional outcome — the same surrogate-versus-clinical-outcome distinction addressed elsewhere in this library (cross-reference ECMO in Severe ARDS protocol's discussion of EOLIA/Bayesian reinterpretation, and DOAC & Anticoagulant Reversal protocol's discussion of ANNEXA-I), and the central organizing theme of this specific protocol.

3–10. [Not Separately Applicable — Cross-Reference Protocol 3]

This protocol addresses adjunctive pharmacotherapy evidence specifically; the full stabilization checklist, diagnostic framework, severity grading (Hunt and Hess/WFNS/modified Fisher), and DCI evaluation/treatment algorithm are comprehensively covered in Protocol 3 (Subarachnoid Hemorrhage) and are not duplicated here.

11. Evidence-Based Management — Three Definitive Trial Programs

Magnesium Sulfate — MASH-2 — A Clean, Unambiguous Negative

• MASH-2 trial (Dorhout Mees et al., Lancet 2012): multicenter RCT, n=606 (with a supporting meta-analysis of 7 total randomized magnesium trials, combined n≈1,204), testing intravenous magnesium sulfate (64 mmol/day) vs. placebo
• Result: intravenous magnesium sulfate did not improve clinical outcome after aneurysmal SAH — a clean, unambiguous negative result reinforced by the pooled meta-analysis of all major magnesium trials
• Practical significance: routine administration of intravenous magnesium to raise serum magnesium above normal is not recommended — among the three agents in this protocol, magnesium's negative result is the least controversial or nuanced; it simply did not work, with no accompanying surrogate-outcome signal of any kind to complicate the interpretation

Statins — STASH — Also a Clean Negative, With a Specific, Instructive Non-Replication Story

• STASH trial (Kirkpatrick et al., Lancet Neurol 2014): international multicenter RCT, n≈803, simvastatin 40 mg daily vs. placebo, started within 96 hours of ictus, continued up to 21 days
• Result: no benefit on the primary ordinal modified Rankin Scale analysis (common OR 0.97, 95% CI 0.75–1.25, p=0.803); 6-month mortality nearly identical (10% simvastatin vs. 9% placebo, p=0.592); no safety concerns identified
• Why this specific negative result was genuinely surprising to the field: STASH followed two smaller, earlier positive pilot trials using simvastatin 80 mg daily (a higher dose than STASH's 40 mg) that had shown promising vasospasm-reduction signals — this is structurally the same single-center/pilot-trial-enthusiasm-not-replicated-at-scale pattern seen repeatedly elsewhere in this library (cross-reference the Glycemic Control protocol's Leuven-to-NICE-SUGAR arc, and Balanced Crystalloids protocol's SMART-to-BaSICS/PLUS arc)
• A specific, unresolved dose-response question left open by STASH: because STASH tested only the 40 mg dose, it did not directly test whether the higher 80 mg dose used in the earlier positive pilots might have shown benefit — an independent, separately-conducted high-dose simvastatin (80 mg) trial (Wong et al.) was specifically designed to test this question, reflecting genuine, still-unresolved uncertainty about whether STASH's negative result reflects statins' mechanism failing entirely, or specifically an inadequate dose
• Independent editorial commentary raised a further, genuinely important methodological point: STASH and the companion high-dose trial were not designed to evaluate neurocognitive outcomes specifically, and the treatment duration (2–3 weeks) may not adequately test statins' plausible longer-term neuroprotective mechanism — the editorial explicitly cautioned against concluding statins have "no role whatsoever" based on STASH alone, while still accepting the trial's clear conclusion that routine acute-phase simvastatin is not supported

Clazosentan — The CONSCIOUS Trial Series — A Genuine, Instructive Surrogate-vs-Clinical-Outcome Dissociation

• CONSCIOUS-1 (Macdonald et al., Stroke 2008, phase 2 dose-finding): established that clazosentan significantly reduces angiographic vasospasm in a dose-dependent manner — a genuine, successful proof-of-mechanism result that directly motivated the two large phase 3 confirmatory trials below
• CONSCIOUS-2 (Macdonald et al., Lancet Neurol 2011, phase 3, surgical clipping population, n≈1,157): clazosentan 5 mg/h vs. placebo — primary composite endpoint (all-cause mortality, vasospasm-related new cerebral infarcts, DCI, rescue therapy for vasospasm) occurred in 21% (clazosentan) vs. 25% (placebo), a 17% relative risk reduction that did not reach statistical significance (95% CI −4 to 33, p=0.10); poor functional outcome (GOSE ≤4) was numerically more common with clazosentan (29% vs. 25%, p=0.10) — a nonsignificant but directionally unfavorable trend on the functional outcome measure specifically; mortality was identical between groups (6% both, week 12)
• CONSCIOUS-3 (endovascular coiling population, n=577 of a planned 1,500, halted prematurely following CONSCIOUS-2's results): tested two clazosentan doses (5 and 15 mg/h) — the higher 15 mg/h dose significantly reduced the primary composite endpoint (vasospasm-related morbidity/all-cause mortality), but neither dose improved functional outcome (extended Glasgow Outcome Scale) — the trial's own published conclusion states this explicitly and directly: clazosentan reduced the composite vasospasm/mortality endpoint "however, neither dose improved outcome"
• This is a genuinely clean, well-documented example of exactly the surrogate-vs-clinical-outcome dissociation theme this protocol is built around: clazosentan reliably, repeatedly reduces the specific pathophysiological target (angiographic vasospasm) it was designed to reduce — across dose-finding, and two separate phase 3 populations (clipping and coiling) — yet this consistent mechanistic success has not translated into improved patient-centered functional recovery in either phase 3 trial, and CONSCIOUS-2 additionally showed a nonsignificant trend toward worse functional outcome specifically
• Adverse effects: lung complications, anemia, and hypotension were more common with clazosentan across the trial program — real, quantifiable costs accompanying the drug's genuine, repeatedly-demonstrated angiographic efficacy

Practical Synthesis

None of the three most extensively studied adjunctive agents beyond nimodipine — magnesium, statins, or clazosentan — are currently supported for routine use in aneurysmal SAH, despite each having a genuine, biologically coherent mechanistic rationale, and despite clazosentan specifically achieving its intended angiographic/mechanistic target repeatedly across multiple trials. This protocol treats the clazosentan story as the most instructive of the three: it is not simply another failed drug, but a clean demonstration that successfully hitting a specific pathophysiological target (angiographic vasospasm) does not guarantee the clinical outcome benefit that target was originally presumed to mediate — directly analogous to the EOLIA/ECMO and ANNEXA-I discordances addressed elsewhere in this library, though notably without EOLIA's redeeming Bayesian reanalysis — clazosentan's functional-outcome signal trended unfavorable, not merely neutral, in CONSCIOUS-2. The core, guideline-endorsed management of DCI remains nimodipine plus induced hypertension for symptomatic, confirmed vasospasm (cross-reference Protocol 3, Sections 11–12) — none of the three agents addressed in this protocol should be added to that established approach based on current evidence.

12–20. [Organ Support / Consultation / Monitoring / Bundle / Complications / Escalation / Discharge / Documentation — Addressed Collectively]

Given this protocol's focused, negative/mixed-evidence pharmacotherapy scope: none of magnesium, statins, or clazosentan should be initiated as routine adjunctive DCI-prevention or -treatment therapy — cross-reference Protocol 3 for the full, established management framework (nimodipine, euvolemia, induced hypertension, angioplasty) that remains the standard of care. No specific monitoring, complication, or escalation pathway is added by this protocol beyond what Protocol 3 already establishes; the relevant addition is purely to the evidence base informing which adjunctive agents are (and are not) supported.

21. Key Guidelines

• Current AHA/ASA and Neurocritical Care Society guidance (cross-reference Protocol 3, Section 20) does not recommend routine magnesium, statin, or clazosentan therapy, consistent with the negative/mixed trial evidence detailed above

22. Landmark Trials

23. Controversies

• The clazosentan story is arguably the clearest, most repeatedly-replicated surrogate-outcome dissociation in this library: three separate trials (dose-finding plus two phase 3 populations) all confirmed the drug does what it was designed to do (reduce angiographic vasospasm/the composite vasospasm-mortality endpoint), yet functional outcome was never significantly improved and trended unfavorable in CONSCIOUS-2 specifically — this protocol treats this as a genuinely important cautionary example for any adjunctive DCI therapy development going forward: angiographic vasospasm reduction alone should not be treated as an adequate surrogate for patient-centered benefit in future trial design or clinical decision-making.
• STASH's dose question remains genuinely unresolved, not definitively closed: the trial tested only 40 mg after earlier pilots showed promise at 80 mg, and a dedicated high-dose confirmatory trial was specifically designed to address this gap — clinicians should be aware this is an open dose-response question rather than a fully settled "statins don't work at any dose" conclusion, even though STASH itself supports not using the 40 mg regimen routinely.
• Whether longer-duration or neurocognitive-outcome-focused trials might yet find a role for statins was explicitly raised by independent editorial commentary on STASH — a reasonable, still-open question given the trial's specific duration (2–3 weeks) and outcome measures (functional mRS, not neurocognitive testing) may not have adequately captured statins' plausible longer-term mechanism.
• Why did clazosentan's angiographic success not translate to functional benefit? Plausible explanations include: DCI may be a more multifactorial process than angiographic vasospasm alone captures (microvascular dysfunction, cortical spreading depolarization, and early brain injury all contribute independently, per Protocol 3's Section 8 discussion of early brain injury), and clazosentan's own adverse effect profile (pulmonary complications, anemia, hypotension) may have offset any genuine benefit from vasospasm reduction — neither explanation has been definitively confirmed, and this remains a genuinely open mechanistic question in the field.

24. References

1. Dorhout Mees SM, Algra A, Vandertop WP, et al; MASH-2 Study Group. Magnesium for aneurysmal subarachnoid haemorrhage (MASH-2): a randomised placebo-controlled trial. Lancet. 2012;380(9836):44-49.
2. Kirkpatrick PJ, Turner CL, Smith C, Hutchinson PJ, Murray GD; STASH Collaborators. Simvastatin in aneurysmal subarachnoid haemorrhage (STASH): a multicentre randomised phase 3 trial. Lancet Neurol. 2014;13(7):666-675.
3. Macdonald RL. Are statins to be STASHed in subarachnoid haemorrhage? [editorial]. Lancet Neurol. 2014;13(7):639-641.
4. Wong GK, Liang M, Tan H, et al. High-dose simvastatin for aneurysmal subarachnoid hemorrhage: a multicenter, randomized, controlled, double-blind clinical trial protocol. Neurosurgery. 2013;72:840-844.
5. Macdonald RL, Kassell NF, Mayer S, et al. Clazosentan to overcome neurological ischemia and infarction occurring after subarachnoid hemorrhage (CONSCIOUS-1): randomized, double-blind, placebo-controlled phase 2 dose-finding trial. Stroke. 2008;39(11):3015-3021.
6. Macdonald RL, Higashida RT, Keller E, et al. Clazosentan, an endothelin receptor antagonist, in patients with aneurysmal subarachnoid haemorrhage undergoing surgical clipping: a randomised, double-blind, placebo-controlled phase 3 trial (CONSCIOUS-2). Lancet Neurol. 2011;10(7):618-625.
7. Macdonald RL, et al. Randomized trial of clazosentan in patients with aneurysmal subarachnoid hemorrhage undergoing endovascular coiling (CONSCIOUS-3). Stroke. 2012.
8. Network meta-analysis of strategies targeting vasospasm and cerebral ischemia after aneurysmal SAH. 2022.
9. The Washington Manual of Critical Care, 4th ed. 2025 — aneurysmal subarachnoid hemorrhage chapter.
10. Keyrouz SG, Diringer MN. Clinical review: prevention and therapy of vasospasm in subarachnoid hemorrhage. Crit Care. 2007;11:220.