February 6, 2019

EJRC - Cardiac Arrest Prognosis

Automated Quantative Pupillometry, a reliable tool to predict unfavourable outcome in comatose post anoxic patients.


Automated pupillary assessment is of interest in the multi-parametric assessment of comatose patients presenting post-anoxic encephalopathy after cardiac arrest, since Pupillary Light Reflex (PLR) examination is part of ERC-ESICM Guidelines on post resuscitation care with Glasgow motor score, EEG, MRI, SSEP.

This Paper described a new automated tool able to perform a more comprehensive quantitative assessment of PLR with the Neurological Pupil Index (NPi). The latter being a scalar value, based on an algorithm that accounts for several variables – including pupil size, percentage and velocity constriction, dilation latency and velocity – when submitted to a calibrated light stimulation of fixed intensity.

Following STARD Guidelines, the authors report on performance of NPi compared with standard PLR in a prospective international study including comatose cardiac arrest patients evaluated between Day 1 and 3, to predict unfavourable outcome (define as a Glasgow Pittsburg cerebral performance category > 2).

Unaffected by confounders, such as Hypothermia and sedation, early prognostication using NPI evaluation had a significantly higher specificity than Standard PLR, with no false Positive and appears accurate and reliable at Day 1 and 2. At Day 3 these differences were not statistically significant.



This is a large multi-centric prospective study with strong methodology.

Study also shows that an accurate prediction of poor neurological outcome using PLR does not require a complete abolition of pupillary activity. NPI <2 at Day 1 or 2 seems a sufficient criterion.

But it is an observational study, with non-homogenous protocol of withdrawal of life sustaining therapies.

With heterogeneous post-resuscitation care in between centres, standard PLR performance may have been impaired by sedation.

Lack of data concerning standard PLR evaluation.

A Pupillary reflex test is recommended at Day 3 (not 1 or 2) in ERC/ESICM guidelines.   It is possible that such an early evaluation of pupillometry at day 1 and 2 may lead to some kind of self-fulfilling prophecy.



For patients presenting with comatose after cardiac arrest, have a close look at the patient PLR; Indeed, Automated PLR evaluation with NPi, might help to get a reliable early prognostication for an objective multi-para-metre assessment.


This article review was prepared and submitted by Mathieu Serie, CHT Gaston Bourret, Nouméa (New Caledonia) and Paul Abraham, CHU, Lyon (France) on behalf of the ESICM Journal Review Club.


1) ERC ESICM guidelines on post resuscitation care, Resuscitation 95 (2015) 202–222

2) Ann Neurol., Early prediction of coma recovery after cardiac arrest with blinded pupillometry, 2017 Jun;81(6):804-810. doi: 10.1002/ana.24943

Comment on this news