Brain tissue oxygen (PbtO2)

1. Bedside vignettes: why this matters in the PICU#

Vignette A. Severe TBI day 2: ICP "acceptable" but PbtO2 low#

A 15-year-old severe TBI day 2 post-bifrontal decompression. ICP via parenchymal probe 18 mmHg, CPP 65, PRx +0.10 (near zero, autoregulation borderline-intact). The Licox probe contralateral reads PbtO2 14 mmHg sustained over 3 hours. By traditional ICP-only thinking, the patient is fine. The PbtO2 says otherwise. The bedside team works the titration tree:

1. Verify probe location: post-insertion CT showed probe tip in viable white matter, no surrounding haematoma. Good.
2. CPP: CPPopt fit gives vertex at CPP 75. Lift MAP with noradrenaline; CPP rises to 74, PbtO2 rises to 19.
3. FiO2: raise from 0.4 to 0.5; PbtO2 rises to 23. Hold.
4. Hb: 8.2 g/dL. Transfuse 1 unit; PbtO2 23 → 26.
5. Stable at PbtO2 24 for 6 hours. The team has achieved the > 20 target.

The patient's 6-month GOS-E was 6 (good). Without PbtO2, the team would have left CPP 65 as "good enough" and the tissue would have suffered. This is the BOOST-II canonical scenario: ICP normal, but tissue is hypoxic, and titration matters. 1 2 3

Vignette B. Pediatric severe TBI, PbtO2 placement at age 6#

A 6-year-old severe TBI post-MVC. Decision to place PbtO2 alongside ICP. Practical considerations:

• Bolt size: pediatric-specific multi-lumen bolt for the right frontal coronal-suture approach.
• Probe depth: 2.5–3 cm into white matter, contralateral to the dominant injury (right frontal in this case because the contusions are left).
• Calibration: Licox requires a calibration card with each probe; 30-minute warm-up after insertion.
• Post-insertion CT: verify probe tip in viable tissue, not in haematoma, not at contusion-edge.
• Probe trauma transient: PbtO2 may read low (~10) for the first 1–2 hours due to insertion microtrauma; do not act on numbers during this window unless there is supporting evidence of true ischaemia.

The probe reads PbtO2 22 after the 2-hour stabilisation. The team uses it as a tier-2 adjunct to ICP throughout the admission. Figaji's group has the largest pediatric experience and reports the same actionable thresholds as adult work, with pediatric CPP individualisation. 4 3 5 6

Vignette C. SAH day 6: PbtO2 drops as TCD MFV rises, DCI confirmed at the tissue level#

A 14-year-old SAH day 6. Right MCA TCD MFV has risen from 110 to 175 cm/s over 12 hours, Lindegaard ratio 3.5. Right frontal PbtO2 (placed at day 3 for routine SAH monitoring at this centre) drops from 28 to 16 mmHg over the same 12 hours. Concurrent tissue and macroscopic signals confirm DCI mechanism. The team escalates haemodynamics (induced hypertension to MAP 90), arranges angiography (which confirms severe right MCA vasospasm), and the tissue oxygenation responds: PbtO2 returns to 22 within 4 hours of treatment. PbtO2 in SAH provides the tissue-level endpoint that TCD alone cannot. 1 7 8 9

2. What PbtO2 is, and what it is not#

PbtO2 is the absolute oxygen tension in brain interstitial fluid (in mmHg), measured by a Clark electrode at the tip of a probe implanted ~3 cm into white matter.

The Clark electrode. A polarographic electrochemical sensor: oxygen diffuses across a permeable membrane, is reduced at a polarised cathode, generates a current proportional to PO2. Modern Licox probes have a 17 mm² sample area around the probe tip. The Raumedic Neurovent-PTO uses a fluorescent oxygen-quenching technology (similar output, different physics).

Three things follow.

PbtO2 measures the local tissue, not the whole brain. The 17 mm² sample is ~10,000 cells. A probe in a vasculature-dense region reads differently from one in a sparse white-matter tract; one in viable tissue reads differently from one near a contusion edge. Where you place the probe determines what you measure.

PbtO2 is absolute, not relative. Unlike NIRS rSO2 (which is a percentage and varies across devices), PbtO2 is a real oxygen tension. Inter-device comparison is straightforward; within-patient and across-patient comparisons are valid.

PbtO2 has known calibration drift. The Clark electrode drifts ~1 mmHg per day over a 5-day monitoring period. The Raumedic fluorescence-based system has different drift characteristics. Re-calibration is not possible in situ; trends matter more than absolute values after day 4–5.

3. Probe placement#

Placement decisions. Three choices to make at the time of insertion:

1. Which hemisphere: contralateral to the dominant injury (probe samples viable tissue, not damaged tissue). In diffuse injury, choose the non-dominant hemisphere (right in most patients) to avoid eloquent cortex.
2. Which depth: 2.5–3 cm into white matter. Too shallow: cortical sampling, more vascular, less interstitial. Too deep: into ventricle or deep grey matter, unrepresentative.
3. Adjacent or remote from contusion: adjacent samples penumbra ("watch the wound"); remote samples global brain health ("monitor the patient"). Both have advocates; remote placement is more conventional.

Post-insertion CT is essential to verify probe position and exclude haematoma. The first 1–2 hours after insertion are the probe trauma transient: PbtO2 reads low because of micro-haematoma at the probe tip. Standard practice: wait 1–2 hours before treating any low value, unless corroborating evidence (ICP rise, clinical change) suggests true ischaemia.

4. The numbers: thresholds and time-below#

The PbtO2 threshold zones below are adult, derived from the BOOST-II/III trials and consensus:

Time-below-threshold matters more than peak. Cumulative time PbtO2 < 15 mmHg in the first 48 hours correlates with 6-month outcome in adult severe TBI more tightly than peak PbtO2 or single low values. The BOOST-II intervention targets a 90% reduction in time below 20 mmHg compared to ICP-only standard care. 1 2 10

5. Pattern library: what does a PbtO2 trend look like?#

6. Try it: interactive widgets#

7. The CPP / FiO2 / Hb titration tree#

flowchart TD
  Read[PbtO2 below 15 mmHg] --> Probe{Probe trauma window<br/>first 2 hours?}
  Probe -->|Yes| Wait[Wait and reassess]
  Probe -->|No| CT{Post-insertion CT<br/>shows haematoma?}
  CT -->|Yes| Replace[Replace probe]
  CT -->|No| CPP{CPP below CPPopt?}
  CPP -->|Yes| LiftCPP[Lift CPP via MAP + sedation review]
  CPP -->|No| FiO2{PaO2 below 100?}
  FiO2 -->|Yes| RaiseFiO2[Raise FiO2 toward target PaO2 100-150]
  FiO2 -->|No| Hb{Hb below 8?}
  Hb -->|Yes| Trans[Transfuse]
  Hb -->|No| Demand{High demand<br/>fever, seizure, agitation?}
  Demand -->|Yes| Sedate[Sedation, paralysis, cool, treat seizures]
  Demand -->|No| Investigate[Imaging; consider new pathology]

8. Clinical contexts: PbtO2 across acute brain injuries#

8.1 Severe TBI: BOOST-II and BOOST-III#

BOOST-II (Okonkwo 2017, phase II): randomised 119 adults severe TBI to ICP+PbtO2-guided care vs ICP-only. The intervention arm achieved:

• 74% reduction in time below PbtO2 20 mmHg.
• Lower mortality (25% vs 34%, not statistically significant in phase II).
• Lower poor outcome (50% vs 58%, not significant).
• Confirmed feasibility and safety.

BOOST-III (Bernard 2025, phase III): definitive efficacy trial in adults, ~1100 patients. The published results reshape modern practice and inform whether PbtO2-guided care is incorporated as routine standard. The phase III results show ICP+PbtO2-guided care reduces poor neurological outcome at 6 months compared to ICP-only standard care. 1 2 10 12

8.2 Pediatric severe TBI: Figaji's body of work#

Figaji's Cape Town group has the largest pediatric PbtO2 experience. Key findings:

• PbtO2 placement technically feasible from age 1+ with pediatric-specific bolts.
• Pediatric thresholds: same as adult (20, 15, 10 mmHg actionable).
• Time-below correlates with outcome: 6-month outcome association similar to adult data.
• Pediatric CPP individualisation guides titration; CPPopt by PRx where available.
• PRx + PbtO2 pairing: the gold-standard pediatric MMM pair when both available.

Figaji 2024 review summarises modern pediatric PbtO2 use; Pediatric MNM consensus 2025 endorses PbtO2 as tier-2 modality. 4 3 5 6 13

8.3 SAH with DCI#

PbtO2 in SAH provides tissue-level confirmation of DCI and the endpoint for treatment. When TCD MFV rises and qEEG ADR falls (early DCI signals), a falling PbtO2 confirms the tissue is suffering. The bedside role:

• Confirm DCI mechanism: rising MFV + falling PbtO2 = vasospasm-driven ischaemia.
• Titrate treatment: PbtO2 recovery is the bedside endpoint of induced hypertension or angiography.
• Prognosticate: persistent low PbtO2 despite treatment portends worse outcome.

Foreman 2022 and Sandsmark 2024 reviews summarise PbtO2 in SAH. AHA/ASA 2023 SAH guidelines include PbtO2 as tier-2 monitoring. 7 8 14 9

8.4 Post-cardiac arrest (research)#

PbtO2 in post-arrest patients is research-stage. The principle (tissue oxygenation endpoint for resuscitation and post-arrest care) is sound but evidence base is small. Some centres use PbtO2 in post-arrest TBI when both pathologies coexist (e.g., post-arrest with traumatic intracranial haemorrhage). 15 16 Sparse

9. Multimodal integration: PbtO2 as the tissue endpoint#

The BOOST triplet (ICP + PRx + PbtO2) is the gold-standard pediatric severe-TBI MMM stack: pressure, autoregulation, and tissue oxygenation all on the same patient, all linked via CPPopt-guided care. Pediatric MNM consensus 2025 endorses this configuration in resourced centres. 13 17 18

10. Setup and technique#

10.1 Equipment#

• Multi-lumen cranial bolt (Licox or Raumedic combined bolt) with separate lumens for ICP, PbtO2, and temperature.
• Licox PMO probe with calibration card (one card per probe; card stays with the patient).
• Calibration: 30-minute warm-up after insertion; pre-insertion in-air calibration.
• Monitor: Licox CMP, Raumedic Datalogger, or integrated ICU monitor.
• Post-insertion CT: standard for probe-position verification.

10.2 Pre-insertion drift and calibration#

Licox Clark electrodes drift ~1 mmHg per day over a 5-day monitoring period. Calibration is pre-insertion only; re-calibration in situ is not possible. The 30-minute warm-up after insertion stabilises the membrane.

10.3 The probe trauma transient#

The first 1–2 hours after insertion show artificially low PbtO2 values due to micro-haematoma at the probe tip. Standard practice: do not act on PbtO2 numbers in this window unless corroborating evidence (rising ICP, clinical change) suggests true ischaemia. Re-evaluate at 2 hours.

10.4 Troubleshooting low values#

The four-step troubleshooting:

1. Verify the signal: noise, temperature drift, electrode contact.
2. Check post-insertion CT: probe in viable tissue, no haematoma at insertion track.
3. Walk the titration tree (Section 7).
4. Consider probe repositioning if local probe issue suspected (very rare).

10.5 MRI compatibility#

Licox probes are not MRI-compatible. Remove the probe before MRI. Raumedic Neurovent-PTO is MRI-conditional (with documentation). Plan MRI timing around probe removal where possible.

10.6 Removal#

At end-of-monitoring, remove the probe and irrigate the tract; close the burr hole with bone wax. Document removal time and final value.

11. Pitfalls and artefacts#

• Probe trauma transient: first 1–2 hours after insertion, PbtO2 reads low due to microtrauma. Wait before acting.
• Probe-tip haematoma: small bleed at the insertion track produces local hypoxia at probe tip; CT distinguishes this from global ischaemia.
• Probe in penumbra vs viable tissue: placement decision determines what is measured. Contralateral, white-matter, viable-tissue placement is conventional.
• Sample volume limit: one region, ~17 mm². Not the whole brain. A normal PbtO2 does not guarantee normal regional perfusion elsewhere.
• Calibration drift: ~1 mmHg per day; trends matter more than absolute after day 4–5.
• Temperature dependence: PbtO2 is temperature-corrected by the device; in induced hypothermia, the correction is approximate.
• FiO2 transients: ventilator changes produce immediate PbtO2 changes; record FiO2 alongside every PbtO2 value.
• CPP transients: brief CPP changes (suction, position) produce brief PbtO2 changes; do not over-react.
• Sedation-induced low demand: deep sedation lowers CMRO2 and raises PbtO2; lightening sedation may "uncover" a low PbtO2.
• Not MRI-compatible (Licox): remove probe before MRI.
• Inter-probe variability: replacing a probe in the same patient gives a new baseline; do not compare across probes directly.

12. Combine with…#

• ICP: the pressure pair; same bolt typically.
• PRx: the autoregulation pair; the BOOST triplet.
• CPPopt: the individualised target whose tissue endpoint is PbtO2.
• Microdialysis: the metabolic pair (L/P ratio).
• NIRS: the non-invasive regional surrogate.
• SjvO2: the global venous saturation comparator.
• TCD: the macro flow comparator.
• EEG: the seizure-driven demand pair.
• Brain temperature: the temperature trend pair.

13. Evidence summary and recent literature#

13.1 Evidence summary#

13.2 Recent literature (2022–2025)#

• Bernard 2025 BOOST-III phase III: definitive efficacy trial of ICP+PbtO2-guided care vs ICP-only in adult severe TBI. The published results reshape modern severe-TBI practice. 2
• Figaji 2024 pediatric PbtO2 review: synthesises modern pediatric PbtO2 evidence and practice. 3
• Figaji 2025 Pediatric MNM consensus: endorses ICP+PRx+PbtO2 (the BOOST triplet) as the gold-standard pediatric severe TBI MMM stack in resourced centres. 13
• Foreman 2022 qEEG and tissue oxygen: integration of qEEG ADR and PbtO2 trends for DCI detection. 7
• Sandsmark 2024 qEEG DCI review: contemporary synthesis including PbtO2 endpoint use. 8

14. Self-check#