Clinical background
Cerebral perfusion pressure (CPP) is the net pressure gradient driving cerebral blood flow and is a core concept in neurointensive care. In patients with raised intracranial pressure (ICP), above all in severe traumatic brain injury and intracranial haemorrhage, CPP can fall to levels at which cerebral blood flow is no longer maintained, ischaemia develops and secondary brain injury worsens. At the same time, an excessively high CPP is associated with a risk of hyperperfusion, cerebral oedema and expansion of haemorrhage, particularly when cerebrovascular autoregulation is disturbed. The decision the instrument serves is where to set the pressure target: high enough to avoid ischaemia, low enough not to worsen intracranial hypertension. Without a calculated CPP, treatment is guided by MAP and ICP separately, which does not capture their interaction.
Calculating the cerebral perfusion pressure
The mean arterial pressure (MAP) is measured invasively via an arterial catheter. The intracranial pressure (ICP) is measured with an intraparenchymal pressure probe or an external ventricular drain. Both transducers should be zeroed at the same reference point, approximated by the external auditory meatus, which corresponds to the level of the foramen of Monro. If the arterial pressure is zeroed at heart level and the ICP at the level of the auditory meatus, a hydrostatic error of about 10 mmHg arises at 30 degrees of head elevation, systematically overestimating CPP [2].
The derivation cohort consisted of 158 patients with traumatic brain injury and a Glasgow Coma Scale (GCS) of 7 or less, treated at the University of Alabama with a protocol targeting CPP to at least 70 mmHg through volume expansion, cerebrospinal fluid drainage via ventriculostomy, systemic vasopressors (phenylephrine or noradrenaline) and mannitol [1]. The mean CPP was 83±14 mmHg, ICP 27±12 mmHg and MAP 109±14 mmHg. Overall mortality was 29 per cent, and the proportion of favourable outcomes rose from 35 per cent at GCS 3 to 75 per cent at GCS 7, significantly better than all the historical series compared [1].
Interpretation in practice
The 2017 Brain Trauma Foundation guidelines recommend a CPP target of 60 to 70 mmHg in severe traumatic brain injury, with the caveat that it is not established whether this range represents a lower limit or an optimal target [2]. Clinical interpretation must weigh several factors:
| CPP (mmHg) | Interpretation | Clinical action |
|---|---|---|
| < 50 | Critically low; pronounced risk of ischaemia | Raise MAP with volume loading and a vasopressor, lower ICP with CSF or fluid drainage; consider mannitol if the ICP is raised |
| 50–59 | Low; increased risk of ischaemia | Escalate towards a target of 60–70 mmHg; identify and treat the cause of the low MAP or high ICP |
| 60–70 | Guideline target in severe TBI | Maintain; if autoregulation is disturbed, an individually higher target may be justified (CPPopt) |
| > 70 | Raised; autoregulation may be disturbed | With intact autoregulation this is usually tolerated. With disturbed autoregulation and a high ICP: lower the MAP target, avoid excessive vasopressor use, and consider the risk of hyperperfusion and haemorrhage |
A more recent observational study of 809 adult TBI patients found an asymmetric relationship between CPP and the autoregulation index (PRx): even small reductions below the individual optimum (CPPopt) were consistently linked to worse outcomes (OR 1.04, 95% CI 1.02–1.06 per hour-dose below CPPopt, p < 0.001), while CPP levels above CPPopt generally showed no association with worse outcomes [2]. With disturbed autoregulation it therefore appears safer to be too high than too low.
Validation and performance
The concept of CPPopt, defined as the CPP level at which PRx is lowest, has been validated in a European multicentre study (CENTER-TBI) of 224 adult TBI patients from 21 centres. PRx was a significant predictor of mortality in both univariable and multivariable regression analysis adjusted for age, GCS motor score and pupillary status, with an AUC of around 0.67 for predicting mortality [4]. LPRx, a low-resolution variant based on minute-by-minute rather than second-by-second data, was also significant but had lower discrimination than PRx [4].
In the large Cambridge cohort (809 patients recruited between 2002 and 2023), deviations from CPPopt were confirmed to be associated with outcome. The distance between CPPopt and the lower limit of autoregulation (LLA), introduced as the "Lower Limit Margin", was prognostic: patients with a shrinking margin had 45 per cent mortality compared with 18 per cent in those with a widening margin (p = 0.003) [2].
A Swedish cohort from Uppsala (471 patients, 2008 to 2018) showed that older patients (≥ 65 years, n = 129) had a higher PRx (disturbed autoregulation) and a higher CPPopt than younger patients (16–64 years, n = 342) [3]. A high PRx adversely affected outcome in older patients, but to a lesser degree than in younger ones. Unlike younger patients, older patients did not show better outcomes when CPP lay close to CPPopt, which argues against uncritical application of the CPPopt approach in this group [3].
Limitations
CPP requires invasive ICP monitoring and therefore applies only to patients in whom such monitoring is indicated, typically severe traumatic brain injury, intracranial haemorrhage with a risk of raised ICP, and other severe acute brain injury with reduced consciousness. Without ICP monitoring, CPP cannot be calculated.
A fixed CPP target, whether 60, 70 or 80 mmHg, takes no account of individual autoregulatory status. A large randomised trial failed to show that fixed CPP-directed treatment was superior to ICP-directed treatment in severe TBI [4], which has driven the development of individualised CPPopt targets based on PRx. Calculating CPPopt does, however, require high-resolution monitoring equipment and software that is not available in every intensive care unit [4].
Older patients present a particular challenge. Autoregulation is more disturbed, CPPopt is higher and the relationship between proximity to CPPopt and outcome is weaker than in younger adults [3]. A low MAP is particularly dangerous in this group, and strictly avoiding a low systolic pressure may be more important than hitting a specific CPP target [3].
The commonest source of error clinically is incorrect zeroing of the pressure transducers. If the MAP is zeroed at heart level and the ICP at the auditory meatus, the apparent CPP is systematically raised, which may lead to the risk of ischaemia being underestimated.
References
- Rosner MJ, Rosner SD, Johnson AH. Cerebral perfusion pressure: management protocol and clinical results. J Neurosurg. 1995;83(6):949-962. PMID: 7490638
- Bögli SY, Olakorede I, Beqiri E et al. Cerebral perfusion pressure targets after traumatic brain injury: a reappraisal. Crit Care. 2025;29:207. PMID: 40399968
- Lenell S, Svedung Wettervik T, Howells T et al. Cerebrovascular reactivity (PRx) and optimal cerebral perfusion pressure in elderly with traumatic brain injury. Acta Neurochir (Wien). 2024;166(1):62. PMID: 38305993
- Riemann L, Beqiri E, Smielewski P et al. Low-resolution pressure reactivity index and its derived optimal cerebral perfusion pressure in adult traumatic brain injury: a CENTER-TBI study. Crit Care. 2020;24(1):266. PMID: 32456684
