Clinical, electrographic, and imaging predictors of absent somatosensory evoked potentials in comatose survivors of cardiac arrest: A retrospective cohort study.

Background

Absence of cortical somatosensory evoked potentials (SSEPs) in comatose survivors of cardiac arrest is a proven predictor of poor neurological outcome, yet this test is not widely available or used. The correlation between absent cortical SSEPs and other clinical, electrophysiologic, and electroencephalographic (EEG) markers in this population remains incompletely defined.

Methods

We conducted a retrospective cohort study of comatose adult patients after cardiac arrest who underwent SSEPs and brain MRI, collecting clinical, neurologic, MRI, and EEG data. We classified MRI findings as cortical, subcortical, or combined. We compared variables by SSEP status using appropriate statistical tests, and logistic regression identified predictors of bilateral absence of N20 responses. We assessed model performance using the AUC.

Results

We included 76 comatose patients after cardiac arrest. Absence of SSEPs was associated with higher neuron-specific enolase, abnormal brainstem reflexes, upward gaze, extensor or absent motor responses, and generalized myoclonus. Combined cortical-subcortical MRI involvement and absence of EEG background reactivity were also more frequent. In multivariable analysis, extensor or absent motor response (OR 7.78), absent EEG reactivity (OR 5.98), and combined MRI involvement (OR 8.52) were independently associated with absent SSEPs. The model showed good discrimination (AUC 0.804), improving with addition of sex (AUC 0.866).

Conclusion

Absence of EEG background reactivity, extensor or absent motor response, and combined cortical-subcortical MRI abnormalities are independently associated with absent SSEPs. These widely available markers may provide complementary prognostic information when SSEP testing is unavailable and contribute to multimodal prognostication in comatose survivors of cardiac arrest.