Cerebral Perfusion Pressure (CPP) Calculator
Cerebral Perfusion Pressure (CPP) Calculator
CPP is mean arterial pressure minus intracranial pressure, and the arithmetic is the easy part. The MAP has to be referenced to the foramen of Monro: a transducer at heart level reads 0.74 mmHg high per centimetre of height difference.
Cerebral perfusion pressure
MAP minus ICPMAP 85 mmHg, ICP 18 mmHg, arterial transducer already levelled at the foramen of Monro
Formula
with MAP referenced to the foramen of Monro: MAPbrain = MAPmeasured − 0.74 × h, h being the transducer height above that level in cm
Brain Trauma Foundation target 60–70 mmHg · ICP treatment threshold 22 mmHg
- MAP and ICP
- both in mmHg, both measured at the level of the foramen of Monro. That is the whole difficulty: the systemic circulation is transduced at the phlebostatic axis by convention, and the brain is not there whenever the bed is head-up
- 0.74 mmHg per cm
- the hydrostatic weight of the fluid column between the two levels. Reinstrup and colleagues give 0.074 mmHg per millimetre; the same figure falls out of ρgh with saline at 1005 kg/m³ (0.74) and blood at about 1055 kg/m³ (0.78). At 30 degrees head-up the tragus sits roughly 13 to 18 cm above the phlebostatic axis, so a transducer at the heart overstates the pressure reaching the brain by about 10 to 13 mmHg, and the reported total spread is up to 15 mmHg — a quarter of the target range
- 60 to 70 mmHg
- the Brain Trauma Foundation fourth-edition Level IIB target, derived in severe traumatic brain injury. The guideline is explicit that “whether 60 or 70 mm Hg is the minimum optimal CPP threshold is unclear and may depend upon the autoregulatory status of the patient”
- what CPP is not
- cerebral blood flow. Flow is CPP divided by cerebrovascular resistance, which this calculation cannot see — vasospasm, lost autoregulation, a gradient across a mass lesion, venous outflow obstruction. An adequate CPP is necessary for adequate perfusion, not sufficient
Worked example
MAP 85 mmHg, ICP 18 mmHg, arterial transducer already levelled at the foramen of Monro
No height difference, so the corrected MAP is 85 − 0.74 × 0 = 85 mmHg
85 − 18 = 67.0 mmHg, inside the 60 to 70 mmHg target range
Now raise the bed to 30 degrees and move the transducer nowhere, which puts the tragus about 20 cm above it. The correction is 0.74 × 20 = 14.8 mmHg and the true CPP is 52.2 mmHg — below the target range, with nothing on the monitor having changed
The two inputs move the answer by exactly one mmHg each, independently: an ICP of 19 gives 66.0, and a MAP of 84 gives the same 66.0. There is no weighting in this formula
At MAP 70 and ICP 25 the answer is 45.0 mmHg; at MAP 90 and ICP 10 it is 80.0. Both are the same subtraction, and neither says anything about cerebral blood flow
What the guideline actually says
| Variable | Recommendation | Level |
|---|---|---|
| ICP | “Treating ICP above 22 mm Hg is recommended because values above this level are associated with increased mortality.” | IIB |
| CPP | “The recommended target cerebral perfusion pressure (CPP) value for survival and favorable outcomes is between 60 and 70 mm Hg.” | IIB |
| CPP | “Avoiding aggressive attempts to maintain CPP above 70 mm Hg with fluids and pressors may be considered because of the risk of adult respiratory failure.” | III |
What a mis-levelled arterial transducer costs
| Transducer above the foramen of Monro | Error in MAP and CPP | A displayed CPP of 67 is really |
|---|---|---|
| 0 cm (levelled at the tragus) | 0 mmHg | 67.0 mmHg |
| 10 cm | 7.4 mmHg | 59.6 mmHg |
| 20 cm (about 30 degrees head-up) | 14.8 mmHg | 52.2 mmHg |
| 25 cm | 18.5 mmHg | 48.5 mmHg |
The arithmetic is trivial; the reference level is not
Cerebral perfusion pressure is the gradient driving blood through the brain: mean arterial pressure at the inflow minus the pressure the brain sits in. There is no coefficient in the definition, and one mmHg of intracranial pressure costs exactly one mmHg of CPP. That simplicity is why the number is calculated at every bedside in neurocritical care, and also why almost all the error in it lives in the measurements rather than the sum.
The reference level is routinely got wrong. For the gradient to mean anything the arterial pressure must be measured at the same height as the intracranial pressure — the foramen of Monro, in practice the tragus. Systemic monitoring transduces at the phlebostatic axis by convention, and the moment the bed goes head-up those levels separate. A fluid column exerts about 0.74 mmHg per centimetre, so at 30 degrees the gap of roughly 13 to 18 centimetres inflates the MAP arriving at the brain by about 10 to 13 mmHg. Published work puts the total variation attributable to transducer placement and position at up to 15 mmHg — a quarter of the entire 60 to 70 mmHg target range, in one consistent direction, with nothing on the monitor to flag it.
The targets come from severe traumatic brain injury. The Brain Trauma Foundation’s fourth edition recommends a CPP of 60 to 70 mmHg at Level IIB and treating an ICP above 22 mmHg at the same level, while noting that “whether 60 or 70 mm Hg is the minimum optimal CPP threshold is unclear and may depend upon the autoregulatory status of the patient”. That caveat is the important one: a brain with intact pressure autoregulation holds flow roughly constant across a wide CPP range, while one that has lost it tracks pressure passively. Two things the number cannot do. It is not a flow, because cerebrovascular resistance is invisible here. And it is a global average that says nothing about an already ischaemic territory. Osmotherapy is monitored by its own laboratory measures: see the calculated serum osmolality and the osmolal gap pages rather than this one. A grade is not a diagnosis and a cohort risk is not this patient’s probability: a stratum in which 72 per cent died tells you about that cohort, not which 72 per cent. Every threshold here comes from a named cohort, and cohorts differ in case mix, era and treatment; where your unit’s protocol differs, it takes precedence.
Frequently asked questions
What is the formula for cerebral perfusion pressure?
CPP = MAP − ICP, both in mmHg. There is no coefficient: 1 mmHg of intracranial pressure removes exactly 1 mmHg of perfusion pressure.
Where should the arterial transducer be levelled for CPP?
At the foramen of Monro — in practice the tragus — not the phlebostatic axis. A fluid column exerts about 0.74 mmHg per centimetre, so a transducer at heart level with the bed at 30 degrees overstates the pressure reaching the brain, and the CPP, by roughly 10 to 13 mmHg.
What CPP should I aim for?
The Brain Trauma Foundation gives 60 to 70 mmHg as a Level IIB target in severe traumatic brain injury, and says in the same breath that whether the floor is 60 or 70 “is unclear and may depend upon the autoregulatory status of the patient”. It is a population range from trauma cohorts, not a prescription, and was not derived in subarachnoid or intracerebral haemorrhage.
Does a normal CPP mean the brain is adequately perfused?
No. Flow is pressure divided by cerebrovascular resistance, and resistance is what this calculation cannot see: vasospasm, a gradient across a mass lesion, impaired venous outflow and lost autoregulation all allow an adequate global CPP alongside regional ischaemia.
Can cerebral perfusion pressure be negative?
Arithmetically yes, when intracranial pressure exceeds mean arterial pressure, and it is survivable only briefly. In practice it is much more often an artefact: an unzeroed or mis-levelled transducer, a damped trace, a drain left open, or an ICP peak read during coughing.
Related calculators
References
- Carney N, Totten AM, O’Reilly C, et al. Guidelines for the management of severe traumatic brain injury, fourth edition. Neurosurgery. 2017;80(1):6–15.
- Reinstrup P, Unnerbäck M, Marklund N, et al. Best zero level for external ICP transducer. Acta Neurochir (Wien). 2019;161(4):635–642.
- A solution to the cerebral perfusion pressure transducer placement conundrum in neurointensive care? The dual transducer. Neurocritical Care; accepted manuscript, University of Cambridge repository item 1810/354288.
Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/
