At today’s stroke conference, we briefly discussed the question of whether large brain infarcts with mass effect are associated with elevations in intracranial pressure (ICP). Several residents opined that midline shift closely correlates with ICP.
It’s been a while since I looked into the primary literature on this, so I dug up a few articles. This study, from Dr. Jeffrey Frank (neurointensivist then at the Cleveland Clinic, now at University of Chicago) found that in 19 patients with clinical deterioration after large infarction, only 5 had elevated ICP. The mean ICP was 13.4 mm Hg. Similarly, the cerebral perfusion pressure (CPP) was diminished in only 2 subjects, with the mean CPP being 75 mm Hg.
Dr. Frank went on to discuss the question of whether ICP-lowering therapies are likely to be of benefit. After reviewing the theoretical issues and the then-extant literature, his conclusion was that there are not. In the following figure he shows what can happen after various ICP-lowering interventions:
In another study, Stefan Schwab, et al. did find that patients’ ICP rose throughout the monitoring period, but this rise was always preceded by clinical signs (depressed level of arousal, pupillary asymmetry). With regard to the correlation of CT evidence of shift and ICP:
Pressure gradients as indirect evidence of raised ICP, such as midline shift and compressed perimesencephalic cisterns, were seen on CT, and could be approved with ICP monitoring. However, even severe midline shift on CT did not necessarily reflect the actual ICP values. In the group of patients with mild midline shift, six patients showed ICP values above 30 mm Hg at time of CT. On the other hand nine patients with severe midline shift demonstrated ICP values below 30 mm Hg.
Both of these studies are rather old, and the neuro-ICU literature has exploded since then. However, I found these papers referenced in Wijdicks’ The Practice of Emergency and Critical Care Neurology, which was published in 2010. I think this is generally a good little text, especially for our future stroke / neuro-ICU folks. However, despite having referenced the literature above in his discussion of ICP, what does Dr. Wijdicks recommend for malignant brain infarcts? The following sequence of interventions (Table 27.2; p. 396 in my hardcover edition):
- 23% saline
- Therapeutic hypothermia
My own take is that there is now excellent evidence in favor of hemicraniectomy, at least when performed within the first 48 hours (we’ll be discussing this further at journal club later this month–any food requests?) and a lot of uncertainty about the value of these other measures. If we’re not going to offer hemicraniectomy, and the family isn’t ready to transition to palliative care, then we and the patient are in a very tough situation.
Khalid, I’m counting on you to sort all of this out during your upcoming fellowship 🙂