Prevalence of clinical electroencephalography findings in stroke patients with delirium

Posted on by

Contributed by Noa Mintz, ScB, Brown University, Providence, RI and Michael Reznik, MD, Department of Critical Care Medicine & Neurology, University of Pittsburgh, Pittsburgh, PA

Background

Patients with acute stroke are at a high risk for delirium, which may occur in up to half of cases depending on the type of stroke.1–4 However, post-stroke delirium presents unique diagnostic challenges due to the potential overlap of stroke-related neurological deficits and the cognitive deficits associated with delirium.

Multiple prior studies have investigated electroencephalographic (EEG) markers of delirium in non-neurological patients.5,6 Typically, delirium has been associated with a generalized slowing of EEG background rhythms, among other findings. Despite the need for objective diagnostic tools for delirium in neurological patients, there have not been prior studies investigating EEG changes associated with delirium in patients with stroke.

To explore the possible utility of EEG in post-stroke delirium, we aimed to determine the prevalence of EEG abnormalities in a cohort of stroke patients on days with and without delirium. Although stroke patients would be expected to have more localized EEG abnormalities in the setting of their focal brain lesions, we hypothesized that the prevalence of generalized EEG abnormalities would be higher during episodes of delirium.

Study Design

We leveraged data from two prospective studies of post-stroke delirium that enrolled patients with acute stroke admitted to the neurocritical care unit or stroke unit at our institution. The first study enrolled 40 patients with intracerebral hemorrhage (ICH),7 while the second study enrolled 140 patients with ICH, subarachnoid hemorrhage (SAH), or acute ischemic stroke (AIS) [manuscript currently under review]. Both studies involved daily delirium assessments performed every afternoon (Monday-Friday) by expert clinicians using DSM-5 criteria. Similar DSM-5 criteria were used for chart review-based methods on weekends and holidays.

We retrospectively analyzed data from patients who underwent EEG monitoring as part of their clinical care, extracting findings from clinical EEG reports corresponding to days of delirium assessment. All EEG reports were produced by attending neurologists and followed guidelines from the American Clinical Neurophysiology Society (ACNS).8-10

We categorized EEG abnormalities including slowing and epileptiform abnormalities (EAs) as generalized or focal according to ACNS guidelines (with generalized patterns defined as bilaterally synchronous and symmetric), then compared the prevalence of these EEG abnormalities between patient days with and without delirium. We also performed subgroup analyses excluding 1) patients receiving anesthetics; 2) patients who required a neurosurgical procedure; 3) SAH patients; and 4) patients with deep lesions.

Results

We identified 58 patients (mean age 67.6 years, 55% female, 84% white, median initial NIHSS score 16) who underwent EEG monitoring during their delirium assessment period, with a total of 192 days of combined EEG and delirium monitoring (88% with delirium [n = 169 days]) Of these patients, two-thirds had ICH, 17% had SAH, and 16% had AIS, with more severe strokes and a higher prevalence of ICH compared to patients who were not included in the study cohort because they did not receive EEGs. All patients who had EEGs had at least one positive delirium assessment during the study window.

Overall, generalized EEG abnormalities were significantly more prevalent on days with delirium. In our primary analysis, generalized slowing was noted on 96% of days with delirium compared to 57% of those without delirium (p = 0.03) and generalized epileptiform abnormalities were observed on 38% of days with delirium compared to 13% of days without delirium (p = 0.03). On the other hand, focal EEG abnormalities (slowing and EAs, typically in the region of the stroke) were reported less frequently on days with delirium, however these differences did not reach statistical significance.

Of note, our main findings remained similar even in subgroup analyses excluding certain patients with a higher risk for generalized EEG abnormalities, such as those receiving anesthetics and those with deep lesions or diffuse SAH.

Conclusions and Future Directions

Our study supports the hypothesis that delirium is associated with diffuse neuronal dysfunction even in patients with acute stroke, as evidenced by a higher prevalence of generalized EEG abnormalities during days with delirium compared to days without delirium. These results suggest that delirium is a clinically distinct entity superimposed on expected stroke findings, and that EEG biomarkers could have utility in the diagnosis of post-stroke delirium. They also support the need for further studies investigating the pathophysiology of delirium in patients with acute stroke.

Several limitations should be noted. First, the subset of eligible patients who underwent EEG monitoring had higher stroke severity, which likely led to sampling and selection bias with an imbalance between delirious and non-delirious days and an overrepresentation of patients with hemorrhagic stroke. Second, since we reviewed clinical EEG reports rather than raw EEG data, we could not perform focused EEG analyses directly corresponding to each delirium assessment, nor could we perform quantitative EEG analyses. Finally, there is the possibility that EEGs from non-delirious days might have been influenced by residual effects from prior delirium episodes.

Future research should involve larger prospective studies with continuous EEG monitoring synchronized to delirium assessments to better characterize EEG markers of delirium in patients with stroke. Additional studies should also include quantitative EEG analyses to better understand neuronal network connectivity patterns associated with post-stroke delirium.

References

  1. Rollo, E. et al. Delirium in acute stroke: A prospective, cross‐sectional, cohort study. Eur. J. Neurol. 28, 1590–1600 (2021).
  2. Shaw, R. et al. Delirium in an Acute Stroke Setting, Occurrence, and Risk Factors. Stroke 50, 3265–3268 (2019).
  3. Kutlubaev, M. A., Bikbulatova, L. F. & Akhmadeeva, L. R. Early diagnosis of delirium in elderly patients with acute stroke. Adv. Gerontol. 6, 60–66 (2016).
  4. McManus, J. et al. The course of delirium in acute stroke. Age Ageing 38, 385–389 (2009).
  5. Boord, M. S. et al. Investigating how electroencephalogram measures associate with delirium: A systematic review. Clin. Neurophysiol. 132, 246–257 (2021).
  6. Kimchi, E. Y. et al. Clinical EEG slowing correlates with delirium severity and predicts poor clinical outcomes. Neurology 93, (2019).
  7. Reznik, M. E. et al. A Pilot Study of the Fluctuating Mental Status Evaluation: A Novel Delirium Screening Tool for Neurocritical Care Patients. Neurocrit. Care 38, 388–394 (2023).
  8. Herman, S. T. et al. Consensus Statement on Continuous EEG in Critically Ill Adults and Children, Part I: Indications. J. Clin. Neurophysiol. 32, 87–95 (2015).
  9. Hirsch, L. J. et al. American Clinical Neurophysiology Society’s Standardized Critical Care EEG Terminology: 2021 Version. J. Clin. Neurophysiol. 38, 1–29 (2021).
  10. Tatum, W. O. et al. American Clinical Neurophysiology Society Guideline 7: Guidelines for EEG Reporting. J. Clin. Neurophysiol. 33, 328–332 (2016).

Suggested Citation

Mintz, Noa and Reznik, Michael. Prevalence of clinical electroencephalography findings in stroke patients with delirium; July 2024, Available at: https://deliriumnetwork.org/clinical-electroencephalography-findings-post-stroke-delirium-patients/ (accessed today’s date)

Posted in Delirium Research.

Leave a Reply