Glasgow Coma Scale In Intubated Patients

You're in the ICU. Think about it: the vent hisses. The monitor beeps. And the neurologist asks for a GCS.

You pause. The patient is intubated. Which means sedated. Maybe paralyzed. And you're supposed to give a number out of 15?

Here's the thing — the Glasgow Coma Scale was never built for this. In practice, not really. But we use it anyway. Also, every shift. Day to day, every handoff. Every transfer note.

And most of us are doing it wrong.

What Is the Glasgow Coma Scale

The GCS is a neurological scoring system. Three components. Eye opening. Verbal response. So naturally, motor response. Total score ranges from 3 to 15. Plus, lower is worse. Simple on paper.

Teasdale and Jennett created it in 1974. In real terms, for trauma patients. In the emergency department. Not for intubated, sedated, critically ill people on mechanical ventilation.

But here we are.

The scale breaks down like this:

Eye Opening (E):

• 4 — Spontaneous
• 3 — To speech
• 2 — To pain
• 1 — None

Verbal Response (V):

• 5 — Oriented
• 4 — Confused
• 3 — Inappropriate words
• 2 — Incomprehensible sounds
• 1 — None

Motor Response (M):

• 6 — Obeys commands
• 5 — Localizes pain
• 4 — Withdraws from pain
• 3 — Abnormal flexion (decorticate)
• 2 — Extension (decerebrate)
• 1 — None

Add them up. That's your GCS And it works..

Except when the patient has a tube between their vocal cords. Then the verbal component breaks entirely Not complicated — just consistent..

The Intubation Problem

An intubated patient cannot speak. Period. So the verbal score defaults to 1 — "none." Some clinicians write "VT" (ventilated tube) or "NT" (not testable). Others just document "1T The details matter here. Took long enough..

But a 1 means something specific in the original scale: no verbal response to pain. Practically speaking, not "can't talk because of a tube. " Those are wildly different clinical pictures.

A patient who mouths words, follows commands, tracks you with their eyes — but has an ETT — gets a GCS of 10T (E4 + V1T + M5). Same score as someone in a deep coma who only extends to pain And that's really what it comes down to..

That's not a rounding error. That's a clinical disaster waiting to happen.

Why It Matters

GCS drives decisions. Big ones.

Neurosurgery consults. ICP monitor placement. Trauma activation levels. Which means family conversations about prognosis. So research enrollment. Quality metrics. Triage protocols.

A drop of two points triggers a CT head in many institutions. A score ≤ 8 gets you intubated — if you aren't already. A score that doesn't improve over 72 hours starts conversations about withdrawal of care Nothing fancy..

So when we fudge the numbers on intubated patients, we're not just being imprecise. We're potentially changing the trajectory of someone's life Easy to understand, harder to ignore..

The Sedation Confounder

Here's what makes it worse: most intubated ICU patients are sedated. Still, propofol. Plus, dexmedetomidine. Worth adding: fentanyl. Maybe paralytics if they're on a vent mode that demands it The details matter here..

You cannot assess a GCS on a paralyzed patient. The motor score is meaningless. Full stop. Worth adding: the eye opening is suppressed. The verbal was never there Worth keeping that in mind..

Yet I've seen "GCS 3T" documented on a patient receiving cisatracurium infusion. That's not a neurological exam. That's a medication list Not complicated — just consistent..

And sedation holds? Great in theory. In practice, you wake a patient up, they get agitated, their ICP spikes, you re-sedate them — and you've lost the window. Or you don't do the hold at all because "they're too unstable.

So we document what we can. We guess. We approximate.

And then we act on those approximations like they're data.

How to Actually Assess GCS in Intubated Patients

You need a systematic approach. Not a guess. Not "they seem awake." Here's what works in practice.

Step 1: Know the Sedation Status

Before you touch the patient, check the drips. Propofol at 50 mcg/kg/min? Dex at 1.4 mcg/kg/hr? Fentanyl 100 mcg/hr? Paralytic in the last 4 hours?

If they're on a paralytic — stop. " That's honest. You cannot assess motor. Document "paralyzed, unable to assess GCS.That's useful.

If they're on sedation but not paralyzed — note the agents and doses. Then decide: is a sedation hold safe? Even so, is it ordered? Has it happened?

No hold = no reliable GCS. But document "sedated, GCS not assessable. " Move on And that's really what it comes down to..

Step 2: Assess Eye Opening First

This is the most reliable component in intubated patients. It's also the one people skip.

Don't just walk in and say "eyes open.But did they open them because you turned on the light? So " That's spontaneous (E4). Day to day, opened the door? Spoke?

Test deliberately:

• Approach quietly. Lights dim. Observe. Eyes open? Think about it: e4. - If closed, speak loudly. "Open your eyes." Eyes open? E3.
• If still closed, apply central pain — trapezius pinch, supraorbital pressure, nail bed pressure. Even so, eyes open? Think about it: e2. - No eye opening to any stimulus? E1.

Document the stimulus used. "E2 to trapezius pinch" tells the next provider infinitely more than "E2."

Step 3: Motor Response — The Real Meat

Motor is the most predictive component. It's also the most messed up.

You need a painful stimulus. On top of that, central pain tests cortical processing. Peripheral pain (nail bed) only tests spinal reflexes. Use central: trapezius pinch, supraorbital pressure, mandibular pressure.

Watch the response. Now, not just "they moved. " What did they do?

Obeys commands (M6): "Squeeze my hand." "Stick out your tongue." "Show me two fingers." Two different commands. Both sides. If they do it — M6. This is huge. An intubated patient who follows commands is not GCS 8. They're GCS 10T or 11T depending on eyes Still holds up..

Localizes pain (M5): You pinch the trapezius. Their hand comes up to your hand — or crosses midline to the stimulus. That's localization. Purposeful. Cortical The details matter here. And it works..

Withdraws (M4): Pulls away from pain. Non-purposeful. No crossing midline. Just flexion at elbow, wrist, fingers.

Abnormal flexion (M3): Decorticate. Slow, stereotyped flexion — adduction of arm, flexion of elbow/wrist/fingers, extension of legs. Cortical-spinal disconnect Still holds up..

Extension (M2): Decerebrate. Arms extend, internally rotate. Wrists pronate. Legs extend. Brainstem pattern.

None (M1): Flaccid. No response to deep central pain Small thing, real impact..

Test both sides. Asymmetry matters. But document it. "M5 right, M4 left" tells a story. "M5" doesn't.

Step 4: The Verbal Component — Stop Pretending

Here's my opinion: stop

Here’s my opinion: stop trying to force a verbal score when the patient is intubated. Instead, treat the “E” portion as a pure observation of eye opening, and assign it a “T” (intubated) modifier. Because of that, the same logic applies to the motor categories—use “M‑T” to flag that the response was elicited in the context of a secured airway. This simple suffix instantly signals to the next clinician that the score is provisional and that the patient’s neurologic exam cannot be fully evaluated That's the whole idea..

When you do need to document speech, be explicit about the method you used. Did the patient cough, moan, or produce any sound in response to a command? Record the exact quality and volume, for example: “M‑T: moans to painful stimulus, decorticate pattern, right‑sided predominance.” Such granularity is far more actionable than a blanket “E1” or “V1” designation.

A practical workflow that many trauma teams now adopt looks like this:

1. Identify the sedation/paralysis status – note agents, doses, and whether a sedation hold has been performed.
2. Assess eye opening – use a stepwise stimulus and write the exact maneuver (e.g., “E3 to verbal command”).
3. Apply central painful stimulus – trapezius pinch or supraorbital pressure, and record the motor tier (M6, M5, M4, M3, M2, M1) with side‑specific notation.
4. Document any vocalization – if present, describe pitch, duration, and whether it approximates words; if absent, state “no intelligible speech.”
5. Assign a provisional GCS‑T score – combine the observed components with the “T” suffix (e.g., “GCS‑T 8T: E2‑T, M5‑T, V‑T”).

By consistently applying these conventions, the score becomes a reliable communication tool rather than a source of ambiguity. It also protects clinicians from over‑interpreting a number that was never meant to capture the full spectrum of neurologic function in an intubated patient.

Conclusion

In the ICU, where every piece of data must be parsed quickly, the Glasgow Coma Scale can still serve a purpose—provided we recognize its blind spots. Stop assigning verbal scores to patients who cannot speak, stop conflating spontaneous eye opening with command‑driven responses, and always annotate the context of your assessment. Day to day, when you document the stimulus, the side, and the quality of the response, you give your colleagues the information they need to make safe, informed decisions. The “T” modifier is not a gimmick; it is a safeguard that preserves the integrity of the score while acknowledging the limits imposed by the airway. Embrace this nuance, and you’ll turn a potentially misleading number into a clear, actionable snapshot of neurologic status.