18.4 Intracranial Hypertension as a Disease State (03.B)

Intracranial Hypertension as a Disease State

Quick Answer: Treat ICP >22 mmHg (BTF 4th edition) with CPP 60–70 mmHg. Ladder: HOB ~30°, neck veins free, sedation, osmotherapy, CSF drain, brief hyperventilation as a bridge, then pentobarbital coma, then decompressive craniectomy. DECRA: early bifrontal DC in diffuse TBI caused more unfavorable outcome. RESCUEicp: last-tier DC cut death (26.9% vs 48.9% at 6 months) but added vegetative and severe-disability survivors. Steroids for tumor vasogenic edema, not for ischemic cytotoxic edema or TBI (CRASH).

Raised ICP is how mass, edema, hydrocephalus, and hyperemia kill brain that might otherwise have been salvageable. Independent OpenExamPrep teaching here covers intracranial hypertension as a disease state listed among Neurocritical care diseases in the ABPN Content Specifications. Monitor insertion technique lives in the neuromonitoring chapter; this section is the crisis physiology and the treatment ladder. This guide is not an ABIM or ABPN product.

Three edemas, three tools

The skull is a box (Monro–Kellie): brain, blood, and CSF. Edema is not one fluid.

Osmotic edema from acute hyponatremia is a fourth pattern treated by careful sodium correction, not by dexamethasone. Herniation from a tumor cyst or hematoma is a surgical volume problem; osmotherapy is a bridge to the OR, not the operation.

Dexamethasone (often 4–10 mg, then a taper matched to the tumor protocol) reduces tumor-related vasogenic edema and can wake a patient with a known metastasis and midline shift enough to complete staging. The same drug does not treat malignant MCA cytotoxic edema. The CRASH trial (Lancet 2004) randomized thousands of TBI patients to corticosteroids versus placebo and found higher death with steroids. BTF 4th edition recommends against steroids for improving outcome or reducing ICP in TBI. If the stem is ischemic swelling or traumatic contusions, steroids are the wrong ladder rung.

The crisis ladder, in order you can defend

Simple physiology first. Head of bed about 30°, neck neutral, cervical collar loosened if the spine is cleared, no kinked internal jugular lines, head not jammed into a ventilator circuit. Venous outflow obstruction is an ICP bolus you can fix with a pillow. Treat fever and seizures that raise CMRO2. Keep CPP in the BTF 60–70 mmHg band: CPP = MAP − ICP. Driving CPP >70 mmHg with heavy fluid and pressors has been tied to more ARDS.

Sedation and ventilation. Propofol or a barbiturate lowers CMRO2 and can drop ICP; propofol infusion syndrome is the dose-and-duration tax. Analgesia so coughing does not spike ICP. Hyperventilation works by hypocapnic arteriolar vasoconstriction and reduced cerebral blood volume. Use it as a minutes-long bridge while mannitol, hypertonic saline, an EVD, or an OR is mobilized. Do not leave PaCO2 at 28 mmHg overnight as prophylaxis. Avoid PaCO2 <25 mmHg. The first 24 hours after TBI are a low-CBF window; prophylactic hypocapnia then is especially unattractive.

Osmotherapy. Mannitol 0.25–1 g/kg IV over about 10–20 minutes. It is an osmotic diuretic: the patient can run dry; watch the osmole gap and renal function; serum osmolality much above ~320 mOsm/kg with a vanishing gap means the drug is not helping. Hypertonic saline (3% weight-based boluses, or 23.4% 30 mL over 5–10 minutes, preferably central) expands volume rather than diuresing it — useful when the patient is hypotensive. Watch sodium so you do not leap 20 mEq/L in a day. Neither agent treats the clot that needs evacuation.

CSF drainage. An external ventricular drain treats interstitial/hydrocephalic pressure and can drop ICP immediately. Over-drainage risks upward herniation in a posterior-fossa mass and subdural hygromas. Lumbar drainage is contraindicated when there is a large posterior-fossa mass or obliterated cisterns.

Pentobarbital coma. A last medical tier after the above: a loading sequence then infusion titrated to ICP control, with continuous EEG (often toward burst-suppression). Expect hypotension (pressors), ileus, myocardial depression, and infection risk. You can no longer examine the cortex. Wean after a period of ICP control rather than treating burst-suppression as a lifestyle.

DECRA versus RESCUEicp — teach the difference, not “craniectomy good/bad”

DECRA (NEJM 2011) randomized early bifrontotemporoparietal decompressive craniectomy in diffuse TBI (no mass lesion requiring evacuation) with ICP above 20 mmHg for 15 minutes in an hour, within 72 hours, still at a relatively early tier. Surgery lowered ICP and ICU days but worsened 6-month GOS-E: unfavorable outcome 70% versus 51%; death was similar (19% vs 18%). Survivors included more vegetative outcomes. Teaching line: early DC for diffuse injury at a modest ICP threshold caused more harm than medical care in that trial.

RESCUEicp (NEJM 2016) randomized last-tier DC (unilateral or bifrontal) for TBI with ICP >25 mmHg for 1–12 hours despite stage 1–2 therapies, including barbiturates as a comparator path. At 6 months, death was 26.9% surgical versus 48.9% medical. Vegetative state was 8.5% vs 2.1%; lower severe disability 21.9% vs 14.4%; upper severe disability 15.4% vs 8.0%. Rates of moderate disability and good recovery were similar. Teaching line: last-tier DC saves lives and increases survival in vegetative and severe-disability states; it does not manufacture a large extra group of good recoveries. Consent conversations must say that out loud.

The two trials are not contradictory once you name timing, ICP threshold, injury pattern (diffuse vs mixed), and what “standard care” already included. Exam stems that quote DECRA to forbid all craniectomy, or quote RESCUEicp to promise independence, are both wrong.

Worked stems

A 28-year-old with diffuse TBI, ICP 21–23 mmHg for brief runs on day 1, no mass lesion: DECRA physiology — do not rush to bifrontal DC as first-tier therapy; treat medically, drain CSF, use osmoles.

A 35-year-old with ICP 28–32 mmHg for hours despite sedation, osmoles, EVD, and barbiturates: RESCUEicp-type last-tier decision — DC may keep the patient alive with a real chance of severe disability; document that trade.

A glioblastoma with vasogenic hemispheric edema: dexamethasone and oncology/neurosurgery, not a CRASH-style steroid lecture aimed at TBI.

A malignant MCA infarct day 2 with low ADC throughout the MCA territory: no dexamethasone for the cytotoxic edema; osmotherapy and, in selected patients ≤60 years, hemicraniectomy evidence from DECIMAL/DESTINY/HAMLET pooled analyses — a different operation and evidence base than DECRA’s bifrontal diffuse-injury trial.

Exam traps

Prophylactic hyperventilation to PaCO2 20 mmHg. Steroids for TBI or ischemic cytotoxic edema. Equating DECRA with RESCUEicp. Treating CPP 90 mmHg as a goal with flood-and-pressors. Mannitol in a hypotensive, dry patient when hypertonic saline would support MAP. Independent practice at /practice/abim-neurocritical-care should make you pick the ladder rung and the trial name that matches the stem.