The initial evaluation of a hemorrhagic stroke patient must document, at minimum, a validated coma and severity score, a cranial CT with the hematoma quantified, coagulation and platelet status, a blood pressure profile, and a dysphagia screen — and the first five must be finished before any anticoagulation reversal is given. The non-negotiable item in every guideline set is a non-contrast head CT read for hemorrhage within minutes of arrival, because intracranial hemorrhage accounts for roughly 10 to 15 percent of all strokes in US registries and changes management within that first hour. What follows is the full documented bundle, in the order a stroke team actually performs it.

The Five-Minute Arrival Bundle: ABCs, Two Scores, and a Glucose Value

Arrival documentation starts with a fixed core. Record airway patency, a Glasgow Coma Scale score written as three separate numbers (eye, verbal, motor) rather than a total, and a National Institutes of Health Stroke Scale score carried out by a certified assessor. The NIHSS is documented even in hemorrhage because it gives the baseline against which deterioration is judged.

Glucose comes next, and it matters more than most checklists imply. The American Heart Association and American Stroke Association 2019 guideline for the early management of acute ischemic stroke advises checking blood glucose in every patient with suspected stroke. Hypoglycemia below 60 mg/dL can mimic a focal deficit and must be corrected before any deficit is attributed to the bleed.

Simultaneously, note the exact last-known-well time, not the time the patient was found. That single timestamp governs every subsequent window calculation, and a documented "found down at 07:00" without a last-known-well estimate is an incomplete record.

The arrival bundle also captures a witnessed seizure, vomiting, or loss of consciousness. Each one predicts aspiration risk and alters imaging urgency. Document them as yes or no with a time, not as free text such as "history of collapse."

Non-Contrast CT First, Then Vascular Imaging: What Gets Read Off the Scan

A non-contrast head CT is the initial imaging study of choice, and the 2019 AHA/ASA guideline endorses it as sufficient to exclude hemorrhage before thrombolysis in ischemic stroke. For hemorrhagic stroke, the same scan answers three questions that belong in the note: where the blood is, how much of it there is, and whether it is still expanding.

Location is written in anatomical terms — basal ganglia, thalamus, lobar, cerebellar, pontine — because lobar location in a patient over 55 raises cerebral amyloid angiopathy as a cause, while deep location points toward hypertension. That distinction affects both prognostication and recurrence counseling.

Volume is quantified. The ABC/2 method, described by Kothari and colleagues in Stroke in 1996, multiplies the largest axial diameter, a perpendicular diameter, and the number of slices times slice thickness, then halves the product. A documented hematoma volume above 30 cm³ is a routine threshold for poor outcome discussion, and intraventricular extension is graded separately.

Vascular imaging is added when the clinical picture or age suggests a structural cause. CT angiography identifies a spot sign within the hematoma, an active-contrast leak that predicts expansion, and excludes a ruptured aneurysm or arteriovenous malformation in younger patients and in lobar bleeds.

Grading Systems That Must Appear in the Chart, With Their Cutoffs

Three scored instruments do most of the documentation work. Each has a numeric threshold a covering clinician can act on, which is why a total score with no components is a documentation failure.

Instrument What it scores Range Actionable cutoff When documented
Glasgow Coma Scale Eye, verbal, motor responses 3–15 ≤8 prompts airway decision On arrival, then hourly
NIH Stroke Scale 11 neurologic domains 0–42 ≥6 associated with worse outcome in ICH cohorts On arrival and after any change
ICH Score (Hemphill) GCS, age, volume, IVH, infratentorial origin 0–6 0 = 0% 30-day mortality; 5 = 88%; 6 = 100% After CT is read
FUNC Score Age, GCS, volume, location 0–11 11 = 82% chance of functional independence At 24 hours, for goals-of-care discussion
mRS Global disability 0–6 Premorbid value anchors prognosis Premorbid estimate on arrival

The ICH Score is the one most commonly required in a complete initial record. Hemphill and colleagues published it in Stroke in 2001, deriving the 30-day mortality figures from a single-hospital cohort of 152 patients. Those percentages describe that derivation cohort, and later external validation cohorts have reported somewhat different calibration, so quote the score and note the source rather than treating 88 percent as a universal constant.

Document premorbid modified Rankin Scale separately. A score of 4 or 5 before the bleed changes the entire goals-of-care conversation, and retrofitting that estimate three days later is unreliable.

Coagulation and Platelet Documentation: The Numbers That Change the Antidote

Every hemorrhagic stroke patient needs a coagulation panel and a platelet count documented before reversal decisions. The minimum set is prothrombin time with international normalized ratio, activated partial thromboplastin time, platelet count, and a medication history covering direct oral anticoagulants.

The INR matters most in warfarin patients. The 2022 AHA/ASA guideline for the management of spontaneous intracerebral hemorrhage recommends rapid reversal of warfarin with a vitamin K-dependent factor concentrate in patients with an INR of 2.0 or higher, and thresholds for action shift with the agent involved.

For direct factor Xa inhibitors, the guideline recommends a specific reversal agent, and for dabigatran, idarucizumab, when the last dose was within a defined interval. The practical documentation point is that the drug name and the last dose time belong in the chart, because "on a blood thinner" cannot be reversed with confidence.

Andexanet alfa carries a labeled warning about thrombotic events, including ischemic stroke and myocardial infarction, and about the need to resume anticoagulation when bleeding is controlled. Note that advisory in the record when the agent is given.

Platelet transfusion is not routine. The 2022 guideline does not support prophylactic platelet transfusion in patients on antiplatelet therapy without a planned neurosurgical procedure, so document the reasoning if platelets are given.

Blood Pressure, Head Position, and Temperature: The Vitals That Get Repeated

Blood pressure management in intracerebral hemorrhage is a documented protocol, not a preference. For patients presenting with a systolic pressure between 150 and 220 mmHg, the 2022 AHA/ASA guideline recommends acute lowering to a target of 140 mmHg, with a range of 130 to 150 mmHg, and advises against dropping below 130 mmHg.

Document the agent chosen, the infusion rate, and the reading immediately before and after each change. Labetalol and nicardipine are the usual first-line choices; the guideline text discusses both and notes that the choice depends on heart rate and local protocol.

Systolic below 130 mmHg is avoided because reduced cerebral perfusion pressure in an already compressed brain can extend injury. That is why a single "BP controlled" line fails as documentation.

Record head-of-bed elevation, temperature, and blood glucose on the same flowsheet. Fever after hemorrhage is associated with worse outcome, and the guideline recommends treating it while investigating a source. Normoglycemia is targeted, with hypoglycemia avoided. These three variables are repeated hourly in most stroke units and should appear as timed entries.

Dysphagia and Aspiration Screening Before the First Swallow

A swallow screen is documented before anything goes by mouth, and the deadline is tight. Stroke patients with dysphagia aspirate silently, and post-hemorrhage vomiting is common, so the risk compounds.

Use a validated bedside tool and name it in the note. The AHA/ASA guideline recommends an organized dysphagia screen before oral intake, and tools such as the Gugging Swallowing Screen assign a graded diet rather than a binary pass or fail. A binary "swallow intact" note without the tool named is weak documentation.

If the screen is failed, document nil by mouth status, the speech-language pathology consult request time, and the plan for nutrition. Repeat the screen when the patient's alertness improves, because a screening result from a somnolent first morning does not define the patient's capacity a week later.

Also document oral care initiation. Aspiration pneumonia prevention bundles in stroke units pair every swallow screen with mouth care, and the initial evaluation is where that order belongs.

Medications and Reversal Agents: Documenting Why, Not Just What

The medication reconciliation in a hemorrhagic stroke is a reversal audit. Document each anticoagulant, antiplatelet, and nonsteroidal anti-inflammatory drug with the last dose time, then document the reversal decision and its justification.

Vitamin K plus prothrombin complex concentrate for warfarin, idarucizumab for dabigatran, and andexanet alfa or four-factor prothrombin complex concentrate for factor Xa inhibitors each require a rationale line. The 2022 guideline supports four-factor PCC when specific agents are unavailable, which is a realistic scenario at hospitals without a neurological intensive care unit.

Protamine for heparin-associated hemorrhage and platelet support after recent antiplatelet exposure with planned surgery complete the table. Each entry names the agent, the dose, the time given, and the lab value that triggered it.

Document the interaction check too. A patient on warfarin who also takes aspirin has two reversal problems, and the note should say so.

Baseline Function, Comorbidities, and Goals of Care at Hour Zero

Premorbid function belongs in the first assessment because it anchors every later conversation. Record the estimated modified Rankin Scale before the event, the living situation, and whether the patient has a documented advance directive or health care proxy.

Comorbidity documentation focuses on the conditions that change hemorrhage behavior: hypertension, prior intracerebral hemorrhage, cerebral amyloid angiopathy, liver disease with coagulopathy, and alcohol use disorder. Each one modifies recurrence risk and, in the case of liver disease, reversal strategy.

Cognitive status before the bleed is separate from the GCS. A patient with baseline dementia who arrives with a GCS of 12 is not the same prognostic case as a previously independent adult with the same score.

Goals-of-care documentation is not a single line. Record who participated in the discussion, what was said about the ICH Score mortality estimate, and what the patient or proxy decided. The guideline recommends early and repeated discussion, and the initial note is where the first iteration is timestamped.

Documenting the Differential: What Else Could This Be?

The initial evaluation must record the alternative diagnoses considered and excluded. Hemorrhagic stroke mimics include hemorrhagic transformation of an ischemic infarct, tumor with hemorrhage, venous sinus thrombosis, and amyloid-related inflammation.

Hemorrhagic transformation of an ischemic stroke is the mimic that changes treatment most. It appears as blood within a vascular territory with surrounding edema, and its presence reframes the timeline and the anticoagulation plan. Document how it was distinguished.

Cerebral venous sinus thrombosis typically produces bilateral or parasagittal hemorrhage with headache out of proportion to the clot. CT venography confirms it, and the initial note should state whether it was considered.

Tumor-associated hemorrhage appears in atypical locations with enhancing components on contrast imaging. If contrast was not given, say so and note the plan.

Worked Documentation Example: Two Patients, Two Records

Two patients arrive the same evening and both have a right basal ganglia hemorrhage. Their initial records differ in ways that matter.

Patient A, age 58, arrives 90 minutes after last known well. GCS is 13. Hematoma measures 4.1 by 3.2 cm across five 5-mm slices, so ABC/2 gives about 32.8 cm³. INR is 1.0, platelets 210,000, systolic 198 mmHg, ICH Score 2. Dysphagia screen performed at hour two with a graded result and a soft diet ordered. Premorbid mRS estimated at 0.

Patient B, age 81, found down with unknown last-known-well. GCS is 9. Hematoma measures 5.0 by 4.0 cm across six 5-mm slices, about 60 cm³, with intraventricular extension. INR is 3.4 on warfarin, systolic 176 mmHg, ICH Score 4. Reversal ordered within 40 minutes of the INR result. Premorbid mRS estimated at 2 from a documented history of falls. Goals-of-care discussion held at hour five with the daughter and documented.

The comparison illustrates why the initial record needs numbers. Patient A's record supports an aggressive plan and a discharge prognosis conversation. Patient B's record, built from the same five data fields plus the family meeting note, supports a different discussion entirely. Neither record is complete without the dysphagia screen timing, the reversal rationale, and the premorbid function estimate.

FAQ

Does every hemorrhagic stroke patient need an MRI in the first hour?

No. Non-contrast CT answers the questions that change management in the first hour, and the 2019 AHA/ASA guideline accepts CT alone to exclude hemorrhage before thrombolysis. MRI adds value later for cavernous malformations, microbleed burden, and amyloid angiopathy diagnosis, typically after the patient is stabilized. Document the CT, then note MRI as a planned step rather than an emergent one.

Is the ICH Score the same as the NIHSS?

No, and mixing them up is a common documentation error. The NIHSS measures neurologic deficit severity across 11 domains and ranges from 0 to 42. The ICH Score estimates 30-day mortality from five variables and ranges from 0 to 6. Both belong in the record, and a note that reports only one of them leaves either the deficit or the prognosis unrecorded.

What happens if the patient's last-known-well time is unknown?

Document the uncertainty explicitly and record the first-found time separately. Unknown onset restricts thrombolysis eligibility considerations and shifts the emphasis to imaging-based assessment. The 2019 guideline discusses wake-up stroke and unknown-onset scenarios with advanced imaging selection, and the initial note should state which pathway was chosen and why.

Can a hemorrhagic stroke patient eat before the dysphagia screen?

Not until a validated screen is documented and passed. Aspiration risk is high after intracerebral hemorrhage because of impaired consciousness, vomiting, and brainstem compression. Order nil by mouth, document the screen request time, and record the speech-language pathology recommendation once it exists. The screen is repeated when alertness improves.

Do antiplatelet drugs need to be reversed in every hemorrhagic stroke?

No. Routine platelet transfusion is not recommended for patients on antiplatelet therapy who are not undergoing a neurosurgical procedure, per the 2022 AHA/ASA guideline. Document which antiplatelet, the last dose, and the surgical plan; when no procedure is planned, the note should say reversal was considered and deferred, with the reasoning.

How much does the initial workup cost, and does it vary by site?

The compute is straightforward even without current prices: non-contrast CT, CT angiography, coagulation panel, platelet count, comprehensive metabolic panel, and a reversal agent. Reversal agents are the dominant cost, and four-factor prothrombin complex concentrate and andexanet alfa differ by orders of magnitude in acquisition price. Confirm current figures with your hospital's pharmacy and imaging departments, because list prices shift with contracting.

Which patients need intracranial pressure monitoring documented?

Monitoring is reserved for patients with a depressed level of consciousness, clinical signs of herniation, or hydrocephalus from intraventricular extension. The 2022 guideline addresses ventricular drainage for obstructive hydrocephalus and does not support routine ICP monitoring in all patients. The initial note should record whether signs of raised pressure were present and what monitoring was initiated or explicitly deferred.

Is a lumbar puncture ever part of the initial evaluation?

Rarely, and never before imaging. CT excludes hemorrhage and mass effect first; a lumbar puncture in the setting of an undiagnosed mass risks herniation. When subarachnoid hemorrhage is suspected and CT is negative, the sequence is documented CT first, then lumbar puncture or CT angiography depending on the time since onset.

This article is general educational information about clinical documentation and does not constitute a diagnosis, treatment plan, or medical advice for any individual patient. Management decisions, including blood pressure targets, reversal agents, and surgical referral, must be made by a locally qualified clinician using current institutional protocols and the patient's full clinical picture. In an emergency, contact local emergency services immediately.