By Halinder S. Mangat, MD
Assistant Professor of Neurology, Weill Cornell Medical College
Dr. Mangat reports no financial relationships relevant to this field of study.
SYNOPSIS: Automated infrared pupillometry holds promise as a quantitative, reproducible measure that aids in determining neurological prognosis after cardiac arrest and coma.
SOURCE: Solari D, Rossetti AO, Carteron L, et al. Early prediction of coma recovery after cardiac arrest with blinded pupillometry. Ann Neurol 2017;81:804-810.
The current standard of prediction of outcome after cardiac arrest uses neurological examination, neurophysiological tests, blood biomarkers, and neuroimaging.1 This process remains limited to defining only patients with grossly poor outcomes. Solari et al examined the additional utility of objective pupillary light reactivity in prediction of recovery of coma after cardiac arrest.2
The study was performed at the intensive care unit of Lausanne University Hospital in Switzerland. Patients were enrolled prospectively if they were alive and remained comatose 48 hours after cardiac arrest. Automated infrared pupillometry was performed using the handheld NeuroLight-Algiscan device, which measures pupillary size and pupillary light reactivity up to 0.05 mm. All measurements were blinded and not used for any clinical decision-making, including withdrawal of care. Most patients underwent therapeutic hypothermia and received sedation-analgesia and pharmacological paralysis. Daily neurological exams were performed in addition to video electroencephalogram (EEG) during hypothermia (day 1) on sedation, and post hypothermia (day 3) off sedation-analgesia. Somatosensory Evoked Potentials (SSEP) and serum neuron-specific enolase (NSE) testing were performed. Outcomes were measured as one-year Cerebral Performance Categories (CPC) score.
Of 103 enrolled patients, 53 died (CPC5). Of survivors, 32 had a good recovery (CPC1), 16 had moderate disability (CPC2), two had severe disability (CPC3), and none was in a vegetative state. At 48 hours, all survivors had ≥ 13% pupillary reactivity (median 20%). Of these, nine had lower reactivity at 24 hours, but this increased to ≥ 13% by 48 hours. Pupil size and sedation/analgesia dose were identical in both groups. This threshold of 13% reactivity had a 100% positive predictive value for survival, with a specificity of 100% and sensitivity of 61%.
Pupillometry reactivity threshold of < 13% was comparable to EEG non-reactivity and absent SSEP N20 in positive predictive value of poor outcome, but was superior to neurological examination. Low pupillary reactivity was strongly correlated with high NSE.
COMMENTARY
This study demonstrated the utility of quantitative measurement of pupillary light reactivity in predicting poor prognosis after cardiac arrest in comatose patients. It was performed prospectively with blinding of study data such that
it did not influence clinical decision-making or affect outcomes, most importantly decisions regarding withdrawal of care. Pupillary light reactivity was ≥ 13% at 48 hours in all survivors, and the lower threshold, which occurred in two patients, was not visible on standard neurological examination.
The determination of an objective threshold of reactivity is a useful tool in prognostication and was equal in predictive value to EEG and SSEPs and superior to a neurological examination. Apart from absent SSEP and elevated NSE levels, all other measures of outcome are subjective and can be prejudiced by clinical judgment. No exact criteria are specified for EEG reactivity. In addition, EEG, SSEP, and NSE testing may not always be available and can be expensive. The ability to use a bedside objective measure of brainstem injury that does not require specialized training to perform and correlates with clinical outcomes is of immense
importance.
However, there are some limitations to the study. This is a single-center study, and while the pupillometry findings were not used to prognosticate, EEG and NSE levels were. In addition, U.S. guidelines for prediction of recovery after cardiac arrest recommend prognostication at 72 hours. The current study performed pupillometry at 48 hours, and it is unknown if these results would be reproducible at 72 hours. Last, pupillometry only predicted survival and not degree of recovery.
In conclusion, this study does provide strong proof-of-concept and pilot data on the use of bedside pupillometry as a strong correlate of brainstem injury and clinical outcome after cardiac arrest, and warrants further investigation in a larger study.
REFERENCES
- Wijdicks EF, Hijdra A, Young GB, et al. Practice parameter: Prediction of outcome in comatose survivors after cardiopulmonary resuscitation (an evidence-based review). Neurology 2006;67:203-210. (reaffirmed October 2009).
- Solari D, Rossetti AO, Carteron L, et al. Early prediction of coma recovery after cardiac arrest with blinded pupillometry. Ann Neurol 2017;81:804-810.