Automated Quantitative Pupillometry in the Critically Ill: A Systematic Review of the Literature

Anisocoria After Direct Light Stimulus is Associated with Poor Outcomes Following Acute Brain Injury

BACKGROUND

Assessing pupil size and reactivity is the standard of care in neurocritically ill patients. Anisocoria observed in critically ill patients often prompts further investigation and treatment. This study explores anisocoria at rest and after light stimulus determined using quantitative pupillometry as a predictor of discharge modified Rankin Scale (mRS) scores.

METHODS

This analysis includes data from an international registry and includes patients with paired (left and right eye) quantitative pupillometry readings linked to discharge mRS scores. Anisocoria was defined as the absolute difference in pupil size using three common cut points (> 0.5 mm, > 1 mm, and > 2 mm). Nonparametric models were constructed to explore patient outcome using three predictors: the presence of anisocoria at rest (in ambient light); the presence of anisocoria after light stimulus; and persistent anisocoria (present both at rest and after light). The primary outcome was discharge mRS score associated with the presence of anisocoria at rest versus after light stimulus using the three commonly defined cut points.

RESULTS

This analysis included 152,905 paired observations from 6,654 patients with a mean age of 57.0 (standard deviation 17.9) years, and a median hospital stay of 5 (interquartile range 3-12) days. The mean admission Glasgow Coma Scale score was 12.7 (standard deviation 3.5), and the median discharge mRS score was 2 (interquartile range 0-4). The ranges for absolute differences in pupil diameters were 0-5.76 mm at rest and 0-6.84 mm after light. Using an anisocoria cut point of > 0.5 mm, patients with anisocoria after light had worse median mRS scores (2 [interquartile range 0-4]) than patients with anisocoria at rest (1 [interquartile range 0-3]; P < .0001). Patients with persistent anisocoria had worse median mRS scores (3 [interquartile range 1-4]) than those without persistent anisocoria (1 [interquartile range 0-3]; P < .0001). Similar findings were observed using a cut point for anisocoria of > 1 mm and > 2 mm.

CONCLUSIONS

Anisocoria after light is a new biomarker that portends worse outcome than anisocoria at rest. After further validation, anisocoria after light should be considered for inclusion as a reported and trended assessment value.

Quantitative Pupillometry for Intracranial Pressure (ICP) Monitoring in Traumatic Brain Injury: A Scoping Review

The neurological examination has remained key for the detection of worsening in neurocritical care patients, particularly after traumatic brain injury (TBI). New-onset, unreactive anisocoria frequently occurs in such situations, triggering aggressive diagnostic and therapeutic measures to address life-threatening elevations in intracranial pressure (ICP). As such, the field needs objective, unbiased, portable, and reliable methods for quickly assessing such pupillary changes. In this area, quantitative pupillometry (QP) proves promising, leveraging the analysis of different pupillary variables to indirectly estimate ICP. Thus, this scoping review seeks to describe the existing evidence for the use of QP in estimating ICP in adult patients with TBI as compared with invasive methods, which are considered the standard practice. This review was conducted in accordance with the Joanna Briggs Institute methodology for scoping reviews, with a main search of PubMed and EMBASE. The search was limited to studies of adult patients with TBI published in any language between 2012 and 2022. Eight studies were included for analysis, with the vast majority being prospective studies conducted in high-income countries. Among QP variables, serial rather than isolated measurements of neurologic pupillary index, constriction velocity, and maximal constriction velocity demonstrated the best correlation with invasive ICP measurement values, particularly in predicting refractory intracranial hypertension. Neurologic pupillary index and ICP also showed an inverse relationship when trends were simultaneously compared. As such, QP, when used repetitively, seems to be a promising tool for noninvasive ICP monitoring in patients with TBI, especially when used in conjunction with other clinical and neuromonitoring data.

Noninvasive Neuromonitoring Modalities in Children Part I: Pupillometry, Near-Infrared Spectroscopy, and Transcranial Doppler Ultrasonography

BACKGROUND

Noninvasive neuromonitoring in critically ill children includes multiple modalities that all intend to improve our understanding of acute and ongoing brain injury.

METHODS

In this article, we review basic methods and devices, applications in clinical care and research, and explore potential future directions for three noninvasive neuromonitoring modalities in the pediatric intensive care unit: automated pupillometry, near-infrared spectroscopy, and transcranial Doppler ultrasonography.

RESULTS

All three technologies are noninvasive, portable, and easily repeatable to allow for serial measurements and trending of data over time. However, a paucity of high-quality data supporting the clinical utility of any of these technologies in critically ill children is currently a major limitation to their widespread application in the pediatric intensive care unit.

CONCLUSIONS

Future prospective multicenter work addressing major knowledge gaps is necessary to advance the field of pediatric noninvasive neuromonitoring.

Neuromonitoring During ECMO Support in Children

Extracorporeal membrane oxygenation is a potentially lifesaving intervention for children with severe cardiac or respiratory failure. It is used with increasing frequency and in increasingly more complex and severe diseases. Neurological injuries are important causes of morbidity and mortality in children treated with extracorporeal membrane oxygenation and include ischemic stroke, intracranial hemorrhage, hypoxic-ischemic injury, and seizures. In this review, we discuss the epidemiology and pathophysiology of neurological injury in patients supported with extracorporeal membrane oxygenation, and we review the current state of knowledge for available modalities of monitoring neurological function in these children. These include structural imaging with computed tomography and ultrasound, cerebral blood flow monitoring with near-infrared spectroscopy and transcranial Doppler ultrasound, and physiological monitoring with electroencephalography and plasma biomarkers. We highlight areas of need and emerging advances that will improve our understanding of neurological injury related to extracorporeal membrane oxygenation and help to reduce the burden of neurological sequelae in these children.

The Neurological Pupil index for outcome prognostication in people with acute brain injury (ORANGE): a prospective, observational, multicentre cohort study

BACKGROUND

Improving the prognostication of acute brain injury is a key element of critical care. Standard assessment includes pupillary light reactivity testing with a hand-held light source, but findings are interpreted subjectively; automated pupillometry might be more precise and reproducible. We aimed to assess the association of the Neurological Pupil index (NPi)-a quantitative measure of pupillary reactivity computed by automated pupillometry-with outcomes of patients with severe non-anoxic acute brain injury.

METHODS

ORANGE is a multicentre, prospective, observational cohort study at 13 hospitals in eight countries in Europe and North America. Patients admitted to the intensive care unit after traumatic brain injury, aneurysmal subarachnoid haemorrhage, or intracerebral haemorrhage were eligible for the study. Patients underwent automated infrared pupillometry assessment every 4 h during the first 7 days after admission to compute NPi, with values ranging from 0 to 5 (with abnormal NPi being <3). The co-primary outcomes of the study were neurological outcome (assessed with the extended Glasgow Outcome Scale [GOSE]) and mortality at 6 months. We used logistic regression to model the association between NPi and poor neurological outcome (GOSE ≤4) at 6 months and Cox regression to model the relation of NPi with 6-month mortality. This study is registered with ClinicalTrials.gov, NCT04490005.

FINDINGS

Between Nov 1, 2020, and May 3, 2022, 514 patients (224 with traumatic brain injury, 139 with aneurysmal subarachnoid haemorrhage, and 151 with intracerebral haemorrhage) were enrolled. The median age of patients was 61 years (IQR 46-71), and the median Glasgow Coma Scale score on admission was 8 (5-11). 40 071 NPi measurements were taken (median 40 per patient [20-50]). The 6-month outcome was assessed in 497 (97%) patients, of whom 160 (32%) patients died, and 241 (47%) patients had at least one recording of abnormal NPi, which was associated with poor neurological outcome (for each 10% increase in the frequency of abnormal NPi, adjusted odds ratio 1·42 [95% CI 1·27-1·64]; p<0·0001) and in-hospital mortality (adjusted hazard ratio 5·58 [95% CI 3·92-7·95]; p<0·0001).

INTERPRETATION

NPi has clinically and statistically significant prognostic value for neurological outcome and mortality after acute brain injury. Simple, automatic, repeat automated pupillometry assessment could improve the continuous monitoring of disease progression and the dynamics of outcome prediction at the bedside.

FUNDING

NeurOptics.

Specific thresholds of quantitative pupillometry parameters predict unfavorable outcome in comatose survivors early after cardiac arrest

Aim

Quantitative pupillometry is the guideline-recommended method for assessing pupillary light reflex for multimodal prognostication in comatose patients resuscitated from out-of-hospital cardiac arrest (OHCA). However, threshold values predicting an unfavorable outcome have been inconsistent across studies; therefore, we aimed to identify specific thresholds for all quantitative pupillometry parameters.

Methods

Comatose post-OHCA patients were consecutively admitted to the cardiac arrest center at Copenhagen University Hospital Rigshospitalet from April 2015 to June 2017. The parameters of quantitatively assessed pupillary light reflex (qPLR), Neurological Pupil index (NPi), average/max constriction velocity (CV/MCV), dilation velocity (DV), and latency of constriction (Lat) were recorded on the first three days after admission. We evaluated the prognostic performance and identified thresholds achieving zero percent false positive rate (0% PFR) for an unfavorable outcome of 90-day Cerebral Performance Category (CPC) 3-5. Treating physicians were blinded for pupillometry results.

Results

Of the 135 post-OHCA patients, the primary outcome occurred for 53 (39%) patients.On any day during hospitalization, a qPLR < 4%, NPi < 2.45, CV < 0.1 mm/s, and an MCV < 0.335 mm/s predicted 90-day unfavorable neurological outcome with 0% FPR (95%CI: 0-0%), with sensitivities of 28% (17-40%), 9% (2-19%), 13% (6-23%), and 17% (8-26%), respectively on day 1.

Conclusion

We found that specific thresholds of all quantitative pupillometry parameters, measured at any time following hospital admission until day 3, predicted a 90-day unfavorable outcome with 0% FPR in comatose patients resuscitated from OHCA. However, at 0% FPR, thresholds resulted in low sensitivity. These findings should be further validated in larger multicenter clinical trials.

Development of reference intervals for pupillometry in healthy dogs

BACKGROUND

Pupillometry, the measurement of pupil size and reactivity to a stimulus, has various uses in both human and veterinary medicine. These reflect autonomic tone, with the potential to assess nociception and emotion. Infrared pupillometry reduces inaccuracies that may occur when the pupillary light reflex is determined subjectively by the examiner. To our knowledge, there are no published studies outlining normal reference intervals for automated pupillometry in dogs.

OBJECTIVE

The objective of this study was to develop de novo automated pupillometry reference intervals from 126 healthy canine eyes.

METHODS

The pupillary light reflex (PLR) was measured with a handheld pupillometer (NeurOptics™ PLR-200™ Pupillometer). Parameters recorded included maximum pupil diameter (MAX), minimum pupil diameter (MIN), percent constriction (CON), latency (LAT), average constriction velocity (ACV), maximum constriction velocity (MCV), average dilation velocity (ADV) and time to 75% pupil diameter recovery (T75). One measurement was obtained for each eye.

RESULTS

The following reference intervals were developed: MAX (6.05-11.30 mm), MIN (3.76-9.44 mm), CON (-37.89 to -9.64 %), LAT (0.11-0.30 s), ACV (-6.39 to -2.63 mm/ s), MCV (-8.45 to -3.75 mm/s), ADV (-0.21-1.77 mm/s), and T75 (0.49-3.20 s).

CLINICAL SIGNIFICANCE

The reference intervals developed in this study are an essential first step to facilitate future research exploring pupillometry as a pain assessment method in dogs.

Superior reproducibility and repeatability in automated quantitative pupillometry compared to standard manual assessment, and quantitative pupillary response parameters present high reliability in critically ill cardiac patients

Background

Quantitative pupillometry is part of multimodal neuroprognostication of comatose patients after out-of-hospital cardiac arrest (OHCA). However, the reproducibility, repeatability, and reliability of quantitative pupillometry in this setting have not been investigated.

Methods

In a prospective blinded validation study, we compared manual and quantitative measurements of pupil size. Observer and device variability for all available parameters are expressed as mean difference (bias), limits of agreement (LoA), and reliability expressed as intraclass correlation coefficients (ICC) with a 95% confidence interval.

Results

Fifty-six unique quadrupled sets of measurement derived from 14 sedated and comatose patients (mean age 70±12 years) were included.

For manually measured pupil size, inter-observer bias was -0.14±0.44 mm, LoA of -1.00 to 0.71 mm, and ICC at 0.92 (0.86–0.95). For quantitative pupillometry, we found bias at 0.03±0.17 mm, LoA of -0.31 to 0.36 mm and ICCs at 0.99. Quantitative pupillometry also yielded lower bias and LoA and higher ICC for intra-observer and inter-device measurements.

Correlation between manual and automated pupillometry was better in larger pupils, and quantitative pupillometry had less variability and higher ICC, when assessing small pupils. Further, observers failed to detect 26% of the quantitatively estimated abnormal reactivity with manual assessment.

We found ICC >0.91 for all quantitative pupillary response parameters (except for latency with ICC 0.81–0.91).

Conclusion

Automated quantitative pupillometry has excellent reliability and twice the reproducibility and repeatability than manual pupillometry. This study further presents novel estimates of variability for all quantitative pupillary response parameters with excellent reliability.

Automated Pupillometry as an Assessment Tool for Intracranial Hemodynamics in Septic Patients

Impaired cerebral autoregulation (CA) may increase the risk of brain hypoperfusion in septic patients. Sepsis dysregulates the autonomic nervous system (ANS), potentially affecting CA. ANS function can be assessed through the pupillary light reflex (PLR). The aim of this prospective, observational study was to investigate the association between CA and PLR in adult septic patients. Transcranial Doppler was used to assess CA and calculate estimated cerebral perfusion pressure (eCPP) and intracranial pressure (eICP). An automated pupillometer (AP) was used to record Neurological Pupil Index (NPi), constriction (CV) and dilation (DV) velocities. The primary outcome was the relationship between AP-derived variables with CA; the secondary outcome was the association between AP-derived variables with eCPP and/or eICP. Among 40 included patients, 21 (53%) had impaired CA, 22 (55%) had low eCPP (<60 mmHg) and 15 (38%) had high eICP (>16 mmHg). DV was lower in patients with impaired CA compared to others; DV predicted impaired CA with area under the curve, AUROC= 0.78 [95% Confidence Interval, CI 0.63−0.94]; DV < 2.2 mm/s had sensitivity 85% and specificity 69% for impaired CA. Patients with low eCPP or high eICP had lower NPi values than others. NPi was correlated with eCPP (r = 0.77, p < 0.01) and eICP (r = −0.87, p < 0.01). Automated pupillometry may play a role to assess brain hemodynamics in septic patients.

Pilot Study of Neurologic Pupil Index as A Predictor of External Ventricular Drain Clamp Trial Failure After Subarachnoid Hemorrhage

BACKGROUND

External ventricular drains (EVDs) provide a temporary egress for cerebrospinal fluid (CSF) in patients with symptomatic hydrocephalus following aneurysmal subarachnoid hemorrhage. Before EVD removal, a wean trial, which involves clamping the EVD, is typically attempted to ensure that CSF self-regulation is achieved. Automated infrared pupillometry (AIP) has been shown to detect early neurologic decline. We sought to explore the use of AIP to detect early EVD clamping trial failure.

METHODS

This prospective observational pilot study enrolled aneurysmal subarachnoid hemorrhage patients before an EVD clamp trial. On initiating the clamp trial, nurses included hourly AIP assessment in documentation. Clamp trial outcome was based on neurologic examination and neuroimaging. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) models were constructed to explore computed tomography (CT) versus AIP as predictors of clamp trial outcome.

RESULTS

Among the 30 subjects enrolled, there were 38 clamping trials and 22 successful EVD removals. CT scan as a predictor of clamp trial was found to have a sensitivity of 68.8% and specificity of 89.5% (PPV = 84.6%, NPV = 77.3%). AIP assessment as a predictor of wean trial outcome was found to have a sensitivity of 58.3% and specificity of 100% (PPV = 100%, NPV = 63.2%).

CONCLUSIONS

The pilot study data support that Neurological Pupil index <3 is a potential indicator of early clamp trial failure, but a CT scan has a higher sensitivity and NPV for predicting successful EVD removal. This finding suggests the benefits of including AIP assessments during clamping trials.

The brain in pediatric critical care: unique aspects of assessment, monitoring, investigations, and follow-up

As survival after pediatric intensive care unit (PICU) admission has improved over recent years, a key focus now is the reduction of morbidities and optimization of quality of life for survivors. Neurologic disorders and direct brain injuries are the reason for 11-16% of admissions to PICU. In addition, many critically ill children are at heightened risk of brain injury and neurodevelopmental difficulties affecting later life, e.g., complex heart disease and premature birth. Hence, assessment, monitoring and protection of the brain, using fundamental principles of neurocritical care, are crucial to the practice of pediatric intensive care medicine. The assessment of brain function, necessary to direct appropriate care, is uniquely challenging amongst children admitted to the PICU. Challenges in assessment arise in children who are unstable, or pharmacologically sedated and muscle relaxed, or who have premorbid abnormality in development. Moreover, the heterogeneity of diseases and ages in PICU patients, means that high caliber evidence is harder to accrue than in adult practice, nonetheless, great progress has been made over recent years. In this 'state of the art' paper about critically ill children, we discuss (1) patient types at risk of brain injury, (2) new standardized clinical assessment tools for age-appropriate, clinical evaluation of brain function, (3) latest evidence related to cranial imaging, non-invasive and invasive monitoring of the brain, (4) the concept of childhood 'post intensive are syndrome' and approaches for neurodevelopmental follow-up. Better understanding of these concepts is vital for taking PICU survivorship to the next level.

Instrumental Evaluation of COVID-19 Related Dysautonomia in Non-Critically-Ill Patients: An Observational, Cross-Sectional Study

Coronavirus disease-19 (COVID-19) is a predominantly respiratory syndrome. Growing reports about a SARS-CoV-2 neurological involvement, including autonomic dysfunction (AD), have been reported, mostly in critically-ill patients, or in the long-COVID syndrome. In this observational, cross-sectional study, we investigated the prevalence of AD in 20 non-critically-ill COVID-19 patients (COVID+ group) in the acute phase of the disease through a composite instrumental evaluation consisting of Sudoscan, automated pupillometry, heart rate variability (HRV), and pulse transit time (PTT). All the parameters were compared to a control group of 20 healthy volunteers (COVID− group). COVID+ group presented higher values of pupillary dilatation velocities, and baseline pupil diameter than COVID− subjects. Moreover, COVID+ patients presented a higher incidence of feet sudomotor dysfunction than COVID− group. No significant differences emerged in HRV and PTT parameters between groups. In this study we observed the occurrence of autonomic dysfunction in the early stage of the disease.