Bispectral Index (BIS)

1. Bedside vignettes: why this matters#

Vignette A. Paralysed PICU patient with refractory ICP#

A 10-year-old with severe TBI, ICP 24 despite tier-2 measures, started on cisatracurium infusion to abolish patient-ventilator dys-synchrony. The COMFORT-B and SBS are now useless. You place the BIS sensor across the forehead. SQI 92, EMG 25 (low), BIS 38. Over the next hour, with midazolam-fentanyl titration, BIS settles at 42 and ICP falls to 16. The number is doing its job: confirming sedation depth in a child whose face cannot signal it. 1 2

Vignette B. Adolescent on ketamine infusion#

A 16-year-old with status epilepticus, third-line on a ketamine infusion. You expect BIS to fall into the 40s. Instead, BIS reads 78, the patient is unresponsive, no clinical movement, pupils equal. The cEEG shows electrographic suppression. Ketamine is the classic BIS paradox: it preserves high-frequency activity that the proprietary algorithm interprets as wakefulness. Treat the cEEG and the clinical exam; BIS is misleading here. 3 4

Vignette C. 3-month-old infant on midazolam#

A 3-month-old post-cardiac surgery on a midazolam-fentanyl infusion plus rocuronium for the first 24 h. The BIS sensor reads 62, SQI 88, EMG 15. The team plans to lighten sedation. Is the infant adequately sedated? The BIS algorithm is not validated below 6 months: cortical maturation, sleep-architecture differences, and the spectral characteristics of immature EEG make the number unreliable. Use the BTF / pediatric sedation scoring fallback, not the BIS, in this age band. 2

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

BIS is a single-channel frontal-EEG-derived sedation index produced by a proprietary algorithm (Aspect Medical Systems, now Medtronic). The algorithm combines three EEG-derived features into a 0–100 ordinal score:

1. Bispectral analysis: a higher-order spectral feature that quantifies phase coupling between frequency components. Coupled, organised activity (awake) gives a high score; decoupled, suppressed activity (deep anaesthesia) gives a low score.
2. Power spectral analysis: the relative contribution of beta vs delta-theta-alpha bands. Awake EEG has more beta; deep anaesthesia has more slow waves.
3. Suppression analysis: the fraction of each epoch with isoelectric activity (the "suppression ratio", SR), which dominates the score when SR > 0.

The exact weights are proprietary; the score is calibrated against a sedation continuum across a large healthy-adult anaesthesia dataset.

What BIS does well#

• Sedation depth in paralysed adults: validated for awareness prevention in general anaesthesia (B-Aware, B-Unaware studies). 5 3
• TIVA titration: continuous numeric feedback during propofol-remifentanil anaesthesia.
• Burst-suppression detection: SR is a reliable marker when present.
• Paralysed PICU sedation titration: the most common modern bedside use.

What BIS cannot do#

• Detect seizures: BIS is not a seizure monitor; rhythmic activity can produce anomalous numbers in either direction.
• Prognosticate after brain injury: BIS is not validated for post-arrest or HIE prognostication.
• Predict awareness in pediatric anaesthesia reliably: the meta-analyses show weak benefit and substantial individual variability. 4
• Be interpreted at face value under ketamine, nitrous oxide, dexmedetomidine, xenon: these agents perturb the bispectrum without matching the clinical sedation depth.
• Be used in neonates and very young infants: cortical maturation makes the algorithm baseline unreliable below 6 months.

3. The sensor: where the strip goes#

The BIS sensor is a single-use adhesive strip with four electrodes:

The sensor crosses the forehead; placement takes under 60 seconds. The sensor expires (the conductive gel dries) and must be replaced every 24 hours.

Skin prep matters: alcohol clean and gentle abrasion. Without prep, impedances rise into the kΩ range and the EMG channel reads spuriously high; that drives the algorithm to produce nonsense.

4. The display: the four numbers to read#

Modern BIS monitors display four numbers and one trend:

Always read all four. A BIS of 65 with SQI 85 and EMG 12 is informative. The same BIS with SQI 30 and EMG 70 is meaningless.

5. The numbers to record: the BIS six-pack#

The chart entry pairs the numeric BIS with the clinical context (drug, paralysis, time since insult); a BIS value out of context is uninterpretable.

6. What is normal? Age-banded baseline considerations#

There is no single "normal" BIS in the PICU; the number is interpreted against the intended sedation depth and the age-typical EEG.

Sources: 2 1. The "deep sedation 40–60" target is the historical anaesthesia value; PICU sedation targets are often lighter (50–70 range) to support neurologic exam, except when treating refractory ICP or status epilepticus.

7. What is abnormal? Pattern library#

Decision tree: what to do with the number#

flowchart TD
  Reading[BIS reading] --> SQI{SQI > 50?}
  SQI -->|No| Reprep[Re-prep sensor; not interpretable]
  SQI -->|Yes| EMGcheck{EMG < 50?}
  EMGcheck -->|No| EMGartefact[Likely EMG inflation; check paralysis depth]
  EMGcheck -->|Yes| Drug{Ketamine or paradoxical agent?}
  Drug -->|Yes| Ignore[Number unreliable; use cEEG/exam]
  Drug -->|No| Age{Age > 6 months?}
  Age -->|No| ClinScore[Use clinical sedation scale; BIS not validated]
  Age -->|Yes| Interpret[Interpret per sedation target]

8. Try it: interactive widget#

9. Management: sedation depth titration with BIS#

BIS does not by itself set the sedation regimen. It provides continuous numeric feedback for titration in the patient whose clinical exam is unavailable (paralysed) or unreliable (deep coma).

9.1 Setting the target#

1. Define the goal: refractory ICP control (deep, BIS 40–50); status epilepticus suppression (burst-suppression, BIS < 30 with SR 30–60%); ventilator-tolerance with paralysis (BIS 50–60); standard PICU sedation off paralysis (use clinical scale).
2. Choose the drug: midazolam, fentanyl, propofol (limit < 4 mg/kg/h in pediatrics; propofol infusion syndrome risk), dexmedetomidine, ketamine (note BIS paradox).
3. Set bounds: a lower limit for awareness risk (typically BIS > 40); an upper limit for over-sedation (typically BIS < 70).
4. Titrate within bounds: small dose adjustments every 15–30 minutes based on the trend, not the single number.

9.2 BIS-guided refractory ICP sedation in severe TBI#

In the paralysed severe-TBI patient on tier-2 measures, sedation depth is one knob in the bundle:

1. Establish baseline BIS at current sedation; document SQI and EMG.
2. If ICP > 20 sustained and current sedation appears light (BIS 60–70), deepen midazolam-fentanyl until BIS 40–50; reassess ICP at 30 minutes.
3. If BIS is already 40–50 and ICP remains high, the next move is not to deepen further; it is to escalate (hyperosmolar, controlled hyperventilation, hypothermia, decompressive craniectomy).
4. If pentobarbital coma is initiated for refractory ICP, the BIS target is burst-suppression with SR ~ 50%, confirmed on cEEG.
5. Document the time-integrated burst-suppression to anticipate slow wake-up after weaning.

9.3 BIS-guided burst-suppression for refractory status epilepticus#

For refractory SE on continuous-infusion midazolam, propofol, or pentobarbital:

1. Target electrographic suppression (typically SR 50–75% on cEEG); BIS will read 10–25.
2. Maintain target for the protocol-specified duration (typically 24–48 h with no clinical seizures).
3. Wean by 25% per 6 hours, tracking BIS and SR.
4. Confirm with cEEG every step; BIS alone does not replace cEEG in SE management. 6 7

10. Clinical contexts: where BIS earns its place#

10.1 Severe TBI on continuous infusion sedation and paralysis#

The strongest pediatric use case. Clinical sedation scores fail under NMBA; BIS provides continuous depth feedback. Pair with cEEG to rule out NCSE in patients whose ICP, autoregulation index, or unexplained metabolic derangement suggests it. 8 9

10.2 Refractory status epilepticus, third-line therapy#

BIS supports titration to a burst-suppression target on continuous infusion midazolam, propofol, or pentobarbital. The endpoint is electrographic (cEEG); BIS is the bedside trend. ESETT and ECLIPSE inform the second-line evidence; third-line endpoints are protocol-driven. 6 7 10

10.3 Pediatric cardiac surgery / ECMO#

In post-cardiac-surgery patients on muscle relaxant and high-dose opioid plus midazolam, BIS supports depth titration during the first 24–48 h. ECMO patients are paralysed, sedated, and at neurological risk; BIS is one channel in the multimodal stack alongside NIRS, aEEG, and cEEG. 11 12

10.4 Pediatric general anaesthesia (awareness prevention)#

The historical use case. Adult evidence for awareness prevention with BIS is moderate (B-Aware, B-Unaware); pediatric evidence is weaker. The Cochrane review concludes that BIS reduces awareness modestly compared with clinical signs alone but is not superior to end-tidal anaesthetic concentration monitoring. 4

10.5 Post-cardiac-arrest (caution)#

BIS is not a prognostic tool in post-cardiac-arrest care. The algorithm calibration assumes healthy cortex; injured cortex produces unreliable numbers. Use BIS solely for sedation depth tracking; SSEP, cEEG, NPI, and clinical exam are the prognostic instruments. 13 14 15

10.6 Neonatal HIE and therapeutic hypothermia (avoid)#

BIS is not validated below 6 months and should not be used in cooled neonates. aEEG and cEEG are the cortical electrophysiology arm of the neonatal HIE bundle. 16 17 18

10.7 Bacterial meningitis / encephalitis#

In the sedated and paralysed child with severe acute encephalitis, BIS supports sedation depth titration alongside aEEG and cEEG. BIS does not detect seizures and does not replace cEEG for NCS screening. 19 20

10.8 Brain-death determination (not a substitute)#

BIS is not part of the World Brain Death Project framework. An isoelectric BIS may corroborate, but the formal test requires multi-channel EEG, defined sensitivity, ECG monitoring, and trained interpretation. 21 22

10.9 DKA cerebral oedema (limited role)#

BIS has no proven role in DKA cerebral oedema monitoring; clinical exam, ICP/CPP physiology, and imaging dominate. If the child is intubated and paralysed for raised ICP, BIS can support sedation titration but does not change the management algorithm. 23 24 25

11. Multimodal integration: BIS in the MMM/MNM stack#

12. Setup and technique#

12.1 Equipment#

• BIS monitor: Medtronic BIS Vista, BIS Complete, or integrated into Philips IntelliVue / GE Carescape.
• BIS sensor: single-use four-electrode adhesive strip; BIS Quattro (adult/pediatric) or BIS Pediatric sensor.
• Skin prep supplies: alcohol pad, NuPrep or similar mild abrasive.

12.2 Placement: 60-second protocol#

1. Clean the forehead with alcohol; gentle abrasion improves contact.
2. Position the sensor with the rectangular pad across the forehead: lateral electrode 1 above the temple at FT9 (left); medial electrode 2 above the right eyebrow; medial electrode 3 above the left eyebrow; lateral electrode 4 at FT10.
3. Press firmly for 5 seconds on each electrode; ensure full conductive-gel contact.
4. Connect the cable to the sensor and to the monitor.
5. Run the impedance check: all four electrodes should read green. Yellow or red means repeat skin prep at that site.
6. Wait 60 seconds for the algorithm to acquire enough data for the first BIS value.

12.3 Trust-the-number checklist#

Before acting on a BIS reading, walk through:

• SQI > 50: algorithm has enough signal
• EMG < 50: frontalis contamination is acceptable
• No paradoxical drug (ketamine, nitrous, xenon, dexmedetomidine high dose)
• Patient age > 6 months
• No active electrographic seizure (rule out with cEEG)
• Sedation regimen documented: drug, infusion rate, last bolus time

If any item fails, the BIS number is not interpretable in isolation. Document the failure, default to clinical scoring (when possible), and add cEEG if the question is electrographic.

12.4 Continuous trending#

• Annotate sedation events: every drug change, every bolus, every paralysis change.
• Annotate clinical events: suction, intubation, weaning, parental visits, painful procedures.
• Annotate position changes: a sensor that has drifted produces SQI drops; re-impedance and re-adhere.
• Replace the sensor every 24 h: gel dries, conductivity falls, EMG contamination rises.

12.5 When to remove the sensor#

• The patient is awake and self-aware; BIS adds no clinical information.
• The clinical exam is reliable (off paralysis, off deep sedation); use the clinical scale instead.
• The BIS is persistently uninterpretable (high EMG, low SQI) and the question cannot be answered by the number.

13. Pitfalls#

• EMG inflation: frontalis muscle EMG, in light sedation or under-paralysis, inflates BIS. High BIS + high EMG = re-check paralysis depth, not "the patient is awake".
• Ketamine paradox: ketamine preserves high-frequency activity that the algorithm interprets as wakefulness; BIS reads 70+ at deep clinical sedation depths. Use clinical exam and cEEG, not BIS, under ketamine.
• Nitrous oxide, xenon, dexmedetomidine: each perturbs the bispectrum without matching the clinical depth.
• Age < 6 months: algorithm not validated; values unreliable.
• Hypothermia: deep hypothermia slows EEG and lowers BIS; do not interpret rewarming-period BIS as light sedation if cortex is recovering.
• Brain injury: HIE, severe TBI, large stroke all alter cortical activity and the algorithm's interpretation; BIS may track sedation poorly in these patients.
• Seizures: rhythmic ictal activity can produce anomalous BIS (high or low); BIS is not a seizure monitor.
• Hypoglycaemia: alters cortical activity; BIS reads low even off sedation.
• NCSE under sedation: BIS can read deeply sedated while cEEG shows ongoing electrographic status.
• Sensor drying / detachment: SQI drops, EMG inflates; replace every 24 h and after any suspicious change.

14. Combine with…#

• Continuous full-montage EEG: the gold standard cortical channel; BIS is its abstraction.
• Amplitude-integrated EEG: bedside cortical electrophysiology trend, complementary in many PICU contexts.
• Pupillometry / NPI: brainstem function alongside cortical sedation depth.
• SSEP: N20 plus cEEG for post-arrest prognosis where BIS does not help.
• Foundations: cerebral metabolism: why hypothermia lowers BIS.
• Integration: refractory status epilepticus: BIS-supported burst-suppression titration.
• Integration: discordance triage: what to do when BIS and the clinical exam disagree.

15. Evidence summary#

16. Recent literature (2022–2025)#

• Tasker 2023 (pediatric neurocritical care review): places BIS as a tier-2 bedside modality in pediatric neurocritical care, with sedation titration as the primary use case. 9
• Naim 2023 (pediatric brain injury post-cardiac arrest): explicitly cautions against BIS for prognostication, recommends SSEP and cEEG instead. 14
• Pediatric MMM consensus (Figaji 2025): positions BIS in tier-2 multimodal monitoring; recommends pairing with cEEG to detect NCSE that BIS cannot see. 26
• Pediatric sedation guidelines continue to recommend clinical scoring (COMFORT-B, SBS, RASS) as primary; BIS as adjunct under paralysis.
• Awareness prevention literature: Cochrane and follow-on meta-analyses have not changed substantially; BIS reduces awareness modestly versus clinical signs alone, no advantage over end-tidal anaesthetic monitoring in volatile anaesthesia. 4
• Cardiac surgery and ECMO: BIS in pediatric cardiac surgery and ECMO supports sedation depth tracking; outcome data remain limited. 12

17. Self-check#