A hospital-grade milrinone pathway for cerebral vasospasm.

The order set embedded below is a mockup — a browser-based wireframe, not a production build. It is intended to demonstrate how the clinical logic could be implemented inside an electronic order-entry system. The underlying objective is to show that current evidence can be translated into explicit, guardrailed logic at the point of care.

Runs entirely in the browser. The mockup allows a clinician to step through the decision logic: the safety verification gates, selection of a systolic blood pressure floor, capture of the patient’s baseline heart rate, and the choice to layer milrinone onto induced hypertension. The nurse-facing administration instructions recalculate in real time as selections change.

Following aneurysmal subarachnoid hemorrhage (aSAH), the ruptured aneurysm is secured via surgical clipping or endovascular coiling. The immediate post-procedure period is typically uneventful. Between post-bleed days 4 and 14, degradation of the subarachnoid clot drives cerebral vasospasm — most commonly around day 7. Rising transcranial Doppler velocities, a new focal neurological deficit, a GCS drop of 2 or more, or otherwise unexplained clinical deterioration herald delayed cerebral ischemia (DCI), which contributes substantially to long-term morbidity after aSAH.

First-line therapy is induced hypertension — titration of vasopressor support, most often norepinephrine, to an elevated systolic blood pressure target. When induced hypertension alone is inadequate, intravenous milrinone is added as a second-line adjunct for its vasodilator and inotropic effect on the cerebral microcirculation. Persistent neurological deficit despite both warrants interventional neuroradiology consultation for intra-arterial verapamil or balloon angioplasty.

Milrinone for DCI is used off-label but is widely described in the contemporary literature. Because it is nurse-titrated at the bedside, the pathway must be unambiguous in execution and grounded in the available evidence.

Dr. Chehab leads the clinical work — directing the pathway, shaping the decision logic, and driving the proposal forward. The evidence synthesis was a joint effort, drawing on primary literature and reconciling it against the 2023 American Heart Association / American Stroke Association and Neurocritical Care Society guidelines. I built the interactive order-set mockup above, translating the synthesis into concrete bedside logic — safety gates, hemodynamic goals, and the nurse-facing administration instructions — that a clinician can click through before any formal build work begins.

Key design choices, each grounded in the literature and reflected in the mockup above:

• Starting dose 0.5 mcg/kg/min — the dose behind the only controlled outcome signal to date (MILRISPASM, Lakhal 2021).
• Loading bolus omitted by default. If used for symptom recurrence during the weaning phase, 50 mcg/kg over 10 minutes with a systolic blood pressure ≥100 mmHg precondition (Bernier 2021).
• Two safety-verification gates that block order-set signing: the ruptured aneurysm must be secured, and the patient must have no severe aortic stenosis or hypertrophic obstructive cardiomyopathy.
• Three systolic blood pressure floor options (≥160, ≥180, ≥200 mmHg), each tied to a clinical profile, with a separate mean arterial pressure floor (≥90, 100, 110, or 120 mmHg).
• Stacked vasopressor cascade — norepinephrine first-line, phenylephrine second-line for tachyarrhythmia concerns, vasopressin as a third-line adjunct — bundled into the same order-set signature as milrinone.
• Milrinone conditional on the valvular / HOCM gate, capped at 1.5 mcg/kg/min for nurse-driven titration. Escalation beyond this threshold requires a new physician order and concurrent evaluation for interventional rescue.
• Hemodynamic-intolerance logic aligned with MILRISPASM: if systolic blood pressure cannot be maintained at the selected floor despite norepinephrine at 1.5 mcg/kg/min, the milrinone dose is maintained — not increased, not decreased, and the attending is notified. This mirrors the 29% intolerance cohort in the MILRISPASM study.
• Heart-rate trigger anchored to the individual patient: heart rate >100 bpm and an increase of >20 bpm from the pre-titration baseline, rather than a flat threshold, so that patients with elevated baseline heart rates are not incorrectly flagged.
• Renal dosing aligned to the FDA package insert, with CrCl <30 mL/min excluded and an explicit note addressing augmented renal clearance — the more common clinical issue in aSAH (median CrCl 191 mL/min in the MILRISPASM cohort; Morbitzer 2019).
• Vergouwen 2010 consensus DCI definition written into the protocol so bedside clinicians can differentiate DCI from sedation effect, seizure, or other causes of deterioration.
• Four-phase de-escalation protocol — explicit eligibility criteria, milrinone wean step-down schedule, sequenced vasopressor wean, and a defined response sequence for symptom recurrence during weaning.

A protocol document reviewed once and filed on a shared drive does not consistently change bedside practice. Logic embedded into the order-entry workflow — with safety gates, dynamic administration instructions, and patient-specific trigger thresholds — can. The objective of the mockup is to make that translation step visible before any formal build work begins: to demonstrate what it would look like to take a literature synthesis and render it directly into the workflow the bedside team already uses.

If the clinical logic is adopted, the subsequent informatics build is relatively brief. The MiVAR randomized controlled trial (NCT04362527) is expected to re-anchor the evidence base for milrinone in DCI when it reports; this pathway is structured so that it can be updated alongside the literature.

• Lannes et al. 2012. Neurocrit Care. Original Montreal milrinone protocol (n=88).
• Abulhasan et al. 2020. Refractory Montreal cohort (n=322).
• Bernier et al. 2021. J Neurosurg Anesthesiol. Proposed standard protocol.
• Lakhal et al. 2021. MILRISPASM controlled before–after study (n=94).
• Crespy et al. 2019. Intra-arterial plus intravenous versus intravenous-only comparison.
• MiVAR randomized controlled trial (NCT04362527). Phase III, 360 patients, results pending.
• 2023 AHA/ASA guidelines for the management of aSAH.
• 2023 Neurocritical Care Society aSAH guidelines.
• Vergouwen et al. 2010. Consensus definition of DCI.
• Baruch et al. 2001. Bolus pharmacodynamics in heart failure.
• Morbitzer et al. 2019. Augmented renal clearance in hemorrhagic stroke.
• FDA milrinone package insert.