Supratentorial intracerebral hemorrhage volume and other CT variables predict the neurological pupil index

Prediction of neurocritical care intensity through automated infrared pupillometry and transcranial doppler in blunt traumatic brain injury: the NOPE study

PURPOSE

This pilot study aimed to determine the capacity of automated infrared pupillometry (AIP) alone and in combination with transcranial doppler (TCD) on admission to rule out need for intense neuroAQ2 critical care (INCC) in severe traumatic brain injury (TBI).

METHODS

In this observational pilot study clinicians performed AIP and TCD measurements on admission in blunt TBI patients with a Glasgow Coma Score (GCS) < 9 and/or motor score < 6. A Neurological Pupil index (NPi) < 3, Pulsatility Index (PI) > 1,4 or diastolic blood flow velocity (dV) of < 20 cm/s were used to rule out the need for INCC (exceeding the tier 0 Seattle Consensus Conference). The primary outcome was the negative likelihood ratio (nLR) of NPi < 3 alone or in combination with TCD to detect need for INCC.

RESULTS

A total of 69 TBI patients were included from May 2019 to September 2020. Of those, 52/69 (75%) median age was 45 [28-67], median prehospital GCS of 7 [5-8], median Injury Severity Scale of 13.0 [6.5-25.5], median Marshall Score of 4 [3-5], the median Glasgow Outcome Scale at discharge was 3 [1-5]. NPi < 3 was an independent predictor of INCC. NPi demonstrated a nLR of 0,6 (95%CI 0.4-0.9; AUROC, 0.65, 95% CI 0.51-0.79), a combination of NPi and TCD showed a nLR of 0.6 (95% CI 0.4-1.0; AUROC 0.67 95% CI 0.52-0.83) to predict INCC.

CONCLUSION

This pilot study suggests a possible useful contribution of NPi to determine the need for INCC in severe blunt TBI patients on admission.

Machine learning model prediction of 6-month functional outcome in elderly patients with intracerebral hemorrhage

Spontaneous intracerebral hemorrhage (ICH) has an increasing incidence and a worse outcome in elderly patients. The ability to predict the functional outcome in these patients can be helpful in supporting treatment decisions and establishing prognostic expectations. We evaluated the performance of a machine learning (ML) model to predict the 6-month functional status in elderly patients with ICH leveraging the predictive value of the clinical characteristics at hospital admission. Data were extracted by a retrospective multicentric database of patients ≥ 70 years of age consecutively admitted for the management of spontaneous ICH between January 1, 2014 and December 31, 2019. Relevant demographic, clinical, and radiological variables were selected by a feature selection algorithm (Boruta) and used to build a ML model. Outcome was determined according to the Glasgow Outcome Scale (GOS) at 6 months from ICH: dead (GOS 1), poor outcome (GOS 2–3: vegetative status/severe disability), and good outcome (GOS 4–5: moderate disability/good recovery). Ten features were selected by Boruta with the following relative importance order in the ML model: Glasgow Coma Scale, Charlson Comorbidity Index, ICH score, ICH volume, pupillary status, brainstem location, age, anticoagulant/antiplatelet agents, intraventricular hemorrhage, and cerebellar location. Random forest prediction model, evaluated on the hold-out test set, achieved an AUC of 0.96 (0.94–0.98), 0.89 (0.86–0.93), and 0.93 (0.90–0.95) for dead, poor, and good outcome classes, respectively, demonstrating high discriminative ability. A random forest classifier was successfully trained and internally validated to stratify elderly patients with spontaneous ICH into prognostic subclasses. The predictive value is enhanced by the ability of ML model to identify synergy among variables.

7- year Experience with Automated Pupillometry and Direct Integration with the Hospital Electronic Medical Record

INTRODUCTION

Manual pupillary assessments are an integral part of the neurological evaluation in critically ill patients. Automated pupillometry provides reliable, consistent, and accurate measurement of the light response. We established a computer interface that allow for direct download of pupillometer information to our hospital EMR. Here, we report the single center experience.

METHODS

An interface allowing direct download of pupillometer data to our EMR was developed. We then performed a prospective study using an electronic survey distributed to nurse that used pupillometers in 2015, 2018, and 2020 using a 5-point Likert style format to evaluate the acceptance of this implementation.

RESULTS

In 2015, 22 nurses were surveyed with 50% of the respondents citing lack of pupillometers and 41% citing the labor intensity associated with data entry as the reason for the reluctance to use the pupillometer. The number of nurse responses in 2018 increased to 123 with 78% of nurses finding that the direct download to hospital EMR improved the efficiency of their neurological exams. In 2020, 108 nurses responded with similar responses to those in 2018. We added 3 additional questions regarding utility of the pupillometer during the COVID19 pandemic. 58% of nurses were reassured of the neurologic exam when using the pupillometer in lieu of a full exam to limit infectious exposure.

CONCLUSIONS

This is the first report of the implementation of a direct interface to download pupillometer data to the EMR. The positive effect on nursing workflow and documentation of pupillary findings is discussed.

Quantitative pupillometry in patients with traumatic brain injury and loss of consciousness: A prospective pilot study

OBJECTIVE

Loss of consciousness (LOC) is a hallmark feature in Traumatic Brain Injury (TBI), and a strong predictor of outcomes after TBI. The aim of this study was to describe associations between quantitative infrared pupillometry values and LOC, intracranial hypertension, and functional outcomes in patients with TBI.

METHODS

We conducted a prospective study of patients evaluated at a Level 1 trauma center between November 2019 and February 2020. Pupillometry values including the Neurological Pupil Index (NPi), constriction velocity (CV), and dilation velocity (DV) were obtained.

RESULTS

Thirty-six consecutive TBI patients were enrolled. The median (range) age was 48 (range 21-86) years. The mean Glasgow Coma Scale score on arrival was 11.8 (SD = 4.0). DV trichotomized as low (<0.5 mm/s), moderate (0.5-1.0 mm/s), or high (>1.0 mm/s) was significantly associated with LOC (P = .02), and the need for emergent intervention (P < .01). No significant association was observed between LOC and NPi (P = .16); nor between LOC and CV (P = .07).

CONCLUSIONS

Our data suggests that DV, as a discrete variable, is associated with LOC in TBI. Further investigation of the relationship between discrete pupillometric variables and NPi may be valuable to understand the clinical significance of the pupillary light reflex findings in acute TBI.