Sharply Contoured Activity Over a Craniectomy Defect
A 62-year-old man is on day 5 of ICU care after a right decompressive hemicraniectomy for malignant middle cerebral artery infarction. He remains intubated and sedated with propofol, and the team has ordered continuous EEG because of intermittent right-sided facial and arm twitching. A 12 cm frontotemporoparietal bone flap has been removed; the skin flap is tense and bulging, and a bulky dressing covers the right hemisphere. The ICU has only eight functioning scalp electrodes and a limited supply of collodion. The bedside nurse asks you to stay away from the defect so the recording looks cleaner, and the resident wants the study completed in 20 minutes. You note that the right-sided electrodes sit lower than the left because of the dressing.
1. How do you place and secure electrodes given the craniectomy and limited supplies, and what must be documented?
Read the worked answer
Do not simply avoid the defect. The right hemisphere is the symptomatic side; moving electrodes away would degrade coverage of the area most likely to generate the twitching. Use the 10-20 landmarks as far as the dressing allows, measure from nasion-inion and preauricular points, and place Fp2, F4, C4, P4, O2, F8, T4, and T6 at the best achievable positions. If the dressing displaces an electrode, mark the actual location and note the deviation in the record. Use paste rather than collodion near the defect to avoid pressure and flammability, and secure wires with tape so the dressing does not pull them. Keep impedances below 5 kΩ. Document the bone defect, the bulging flap, the modified positions, and the limited montage. Explain to the nurse that a clean-looking but nonstandard recording can hide the very activity being sought; the goal is interpretable coverage, not cosmetic symmetry.
2. The tracing shows 7-8 Hz sharply contoured activity over the right central region. How do you determine whether this is breach rhythm or a seizure?
Read the worked answer
Breach rhythm is a normal variant generated by cortex under a skull defect; it is arciform, sharply contoured, and often 6-9 Hz. A seizure requires evolution in frequency, morphology, and field, plus a plausible onset. Check whether the activity changes with tactile stimulation or contralateral hand movement; breach rhythm typically attenuates or reacts, whereas a seizure does not. Look for phase reversal in a bipolar montage: breach rhythm may show a localized phase reversal but no spread or evolution. Review the raw tracing at 10-15 mm/s and compare with the contralateral side. If uncertain, keep recording, add a closely spaced electrode over the defect, and ask for a brief period without sedation if safe. Do not call it epileptiform based on sharpness alone; document the reactivity and morphology. Tell the resident that sharp contours over a craniectomy are expected and that treatment should not be based on this alone.
3. The resident wants to give a benzodiazepine based on your verbal report. What do you say and do?
Read the worked answer
I would not give a diagnostic opinion outside my role. I would clarify that I have not identified a seizure, that the activity is sharply contoured but reactive and non-evolving, and that the physician needs to review the raw EEG before treating. I would show the resident the tracing, the reactivity testing, and the montage, and offer to extend the recording. I would document the exchange, including that the resident was informed of the uncertainty. If the resident still orders treatment, that is a medical decision; my responsibility is to provide accurate technical information and not to label the pattern as epileptiform without evidence. I would also notify the supervising electroencephalographer.