Page 18

NCS Currents June 2016

TECH CORNER Automated Pupillometry: Neurologist versus Machine By Fawaz Al-Mufti, MD The author has no actual or potential conflict of interest in relation to the topics discussed in this column. We may discuss non-FDA approved devices and “off-label” uses. The NCS and Currents do not endorse any particular device. Neurointensivists focus on the recognition of subtle changes in the neurological examination, systemic derangements, and their interactions with brain physiology. In addition to relying on high clinical acumen, we find ourselves relying on numerous gadgets and ever-evolving technology. Our field has evolved rapidly in the last decade, incorporating new technology such as multi-modality monitoring and advanced imaging techniques. Through the Currents Tech Corner column, we have decided to examine tools frequently used in the Neuro ICU that constitute our technological armamentarium. Some of what will be covered may come across as basic essentials, however, what some may take for granted may be considered experimental by others. Pupil examination is an important clinical parameter for patient monitoring. Current practice is to use a penlight to observe the pupillary light reflex. The result seems to be a subjective measurement, with low precision and reproducibility. Pupillary assessment can be affected by factors related to: the examiner (e.g., visual acuity, level of experience, placement of the light stimulus, and subjectivity of the description of light reaction as “brisk,” “sluggish,” or “nonreactive”), the patient (e.g., iris color, medication-related pupillary changes, or pre-existing ocular, systemic, or neurological disease), and the environment (e.g., amount of ambient light in the examining room). Several papers have described low inter-rater reliability approaching a 30% disagreement on the evaluation of pupillary reactivity in comatose patients. The automated pupillometer was designed to provide consistency and reliability while removing subjectivity from the measurement of the pupil size and reactivity. Handheld pupillometers measure pupillary variables such as size, latency, constriction velocity, and dilation velocity. The Neurological Pupil Index (NPI), is an algorithm developed to provide an objective and quantifiable way for clinicians to rate the pupillary light reflex and is derived from various calculated parameters (e.g., constriction velocity, dilation velocity, and latency) graded on a scale of zero to five. Once a baseline measurement has been obtained, the clinician can track and trend any subtle changes or deterioration in pupillary responsiveness. A score equal to or above three means that the pupil measurement falls within normal behavior (“brisk”). An NPI score below three means the reflex is abnormal – i.e., weaker than a normal pupil response as defined by the NPI model (“sluggish”). The pupillometer measures both pupil size and reactivity by first measuring baseline pupil size during an initial 200 ms latency period, and then emitting a pulse of light lasting 0.8 seconds, and recording the pupillary response to the stimulus using a high temporal resolution video recorder over the subsequent 3.2 seconds. Automated pupillometry may allow clinicians to detect pupillary abnormalities earlier than with the naked eye, thereby allowing them to intervene earlier. In one study of four patients with neurological examination consistent with brain death, they were discovered to have small but detectable pupillary reaction using the pupillometer. Another study assessed the inter-rater reliability for pupillary size and reactivity among neurosurgical attendings, neurosurgical interns, and advanced practice nurses. It found greater error with manual assessment compared with pupillometry and a 39% disagreement rate when manually assessing pupillary reactivity compared to only a 1% rate of disagreement with the pupillometer. This study defined the ‘‘true’’ pupil size by the measurement of the static pupillometer, which could have introduced a bias in favor of automated pupillometry. There are, however, several case reports in the literature that caution against the use of pupillometry as the sole method to determine the absence of the pupillary light reflex, especially when evaluating for the possibility of brain death. In one case, clinical examination clearly revealed pupillary reactivity while the pupillometer reported a non-reactive pupil which was postulated to be due to the very slow rate of constriction of the patient’s pupil. Another limitation is the number of times an automated pupillometer reading could not be obtained (5.9 %). The most common reasons for inability to obtain readings included peri-orbital edema, patient movement (especially in the patient with impaired cognition), cataracts or ocular prosthesis. Automated pupillometry can limit inter-rater differences in assessment of pupillary size and reactivity and is best considered an adjunct to the standard neurological examination. The combined use of the manual examination and pupillometry is advisable to optimize the accuracy of assessment of pupillary reactivity. For additional reading, see: “Reliability of standard pupillometry practice in neurocritical care: an observational, double-blinded study” by Couret et al. in Critical Care 2016; 20:99 “Pupillary reactivity as an early indicator of increased intracranial pressure: the introduction of the neurological pupil index” by Chen et al. in Surg Neurol Int 2011;2:82 “Neurologist versus machine: is the pupillometer better than the naked eye in detecting pupillary reactivity” by Kramer et al. in Neurocrit Care 2014;21:309–311 “Interrater reliability of pupillary assessments” by Olson et al. in Neurocrit Care 2016;24:251–257 18


NCS Currents June 2016
To see the actual publication please follow the link above