| 1. |
- Oddo, Mauro, et al.
(author)
-
Quantitative versus standard pupillary light reflex for early prognostication in comatose cardiac arrest patients : an international prospective multicenter double-blinded study
- 2018
-
In: Intensive Care Medicine. - : Springer Science and Business Media LLC. - 0342-4642 .- 1432-1238. ; 44:12, s. 2102-2111
-
Journal article (peer-reviewed)abstract
- Purpose: To assess the ability of quantitative pupillometry [using the Neurological Pupil index (NPi)] to predict an unfavorable neurological outcome after cardiac arrest (CA). Methods: We performed a prospective international multicenter study (10 centers) in adult comatose CA patients. Quantitative NPi and standard manual pupillary light reflex (sPLR)—blinded to clinicians and outcome assessors—were recorded in parallel from day 1 to 3 after CA. Primary study endpoint was to compare the value of NPi versus sPLR to predict 3-month Cerebral Performance Category (CPC), dichotomized as favorable (CPC 1–2: full recovery or moderate disability) versus unfavorable outcome (CPC 3–5: severe disability, vegetative state, or death). Results: At any time between day 1 and 3, an NPi ≤ 2 (n = 456 patients) had a 51% (95% CI 49–53) negative predictive value and a 100% positive predictive value [PPV; 0% (0–2) false-positive rate], with a 100% (98–100) specificity and 32% (27–38) sensitivity for the prediction of unfavorable outcome. Compared with NPi, sPLR had significantly lower PPV and significantly lower specificity (p < 0.001 at day 1 and 2; p = 0.06 at day 3). The combination of NPi ≤ 2 with bilaterally absent somatosensory evoked potentials (SSEP; n = 188 patients) provided higher sensitivity [58% (49–67) vs. 48% (39–57) for SSEP alone], with comparable specificity [100% (94–100)]. Conclusions: Quantitative NPi had excellent ability to predict an unfavorable outcome from day 1 after CA, with no false positives, and significantly higher specificity than standard manual pupillary examination. The addition of NPi to SSEP increased sensitivity of outcome prediction, while maintaining 100% specificity.
|
|
