Correspondence  |   March 2018
Neurocritical Care Needs Predictive Scores That Succeed at Predicting Failure as Well as They Predict Success
Author Notes
  • Beth Israel Deaconess Medical Center, Boston, Massachusetts (J.P.).
  • (Accepted for publication December 7, 2017.)
    (Accepted for publication December 7, 2017.)×
Article Information
Correspondence   |   March 2018
Neurocritical Care Needs Predictive Scores That Succeed at Predicting Failure as Well as They Predict Success
Anesthesiology 3 2018, Vol.128, 686-687. doi:10.1097/ALN.0000000000002071
Anesthesiology 3 2018, Vol.128, 686-687. doi:10.1097/ALN.0000000000002071
To the Editor:
In the August 2017 issue of Anesthesiology, Asehnoune et al. report their derivation of a novel bedside scoring system to predict extubation success in the intubated brain-injured patient.1  Many brain-injured patients are likely exposed to excess ventilated days because they do not meet extubation criteria originally established in general intensive care unit (ICU) populations.2  Careful consideration is required, however, before routinely utilizing new extubation prognostication scores. Although the VISAGE (visual pursuit, swallowing, age, Glasgow coma scale for extubation) score performs well at predicting extubation success based on favorable neurologic indicators, it does not adequately predict which patients will fail extubation due to neurologic dysfunction.
Recovery of arousal and airway protective reflexes after neurologic injury often is slow, and a subset of patients will benefit from early tracheostomy without an extubation attempt. The VISAGE score poorly discriminates extubation success among patients with low scores. Based on this model, a patient under 40 yr old without visual pursuit or swallowing efforts, and with a Glasgow coma scale less than 10, would have an almost 60% chance of extubation success. Barring a prediction of rapid neurologic improvement or barriers to safe reintubation, we believe that this individual should undergo a trial extubation. We are concerned that adoption of a scoring system with explicit or perceived cut-points would lead to such patients remaining intubated longer than necessary. A similar problem arises from the predictive score introduced in Anesthesiology earlier this year by Godet et al.3  Although their regression-based score has a clear inflection point, fully one third of patients below this score were successfully extubated. At the suggested cut-point, their score falls short of the degree of negative predictive value originally reported for the Rapid Shallow Breathing Index (RSBI) in a general ICU population.4  The negative predictive value for the VISAGE score at a cut-point of 3 performs even worse.
Timely extubation of all ICU patients, including those with brain injury, helps prevent ventilator-associated complications. Although our colleagues highlight that brain-injured patients can be safely extubated, we caution against rigorously applying these scores due to the possibility of excess mechanical ventilation for patients who score poorly. Extubation failure and reintubation is certainly not without risk and is predictive of worse outcomes, though causality has not been established.1,5  Further development of scoring models with improved negative predictive values is needed to identify patients who should truly forgo trial extubation. Until these risks are further quantified, and such a tool is developed, the neurocritical care intensivist will necessarily have to tolerate and manage higher reintubation rates than those seen in a general ICU population.
Competing Interests
The authors declare no competing interests.
Johann Patlak, M.D., Shahzad Shaefi, M.D., Lauren Buhl, M.D., Ph.D., Myles D. Boone, M.D. Beth Israel Deaconess Medical Center, Boston, Massachusetts (J.P.).
Asehnoune, K, Seguin, P, Lasocki, S, Roquilly, A, Delater, A, Gros, A, Denou, F, Mahé, PJ, Nesseler, N, Demeure-Dit-Latte, D, Launey, Y, Lakhal, K, Rozec, B, Mallédant, Y, Sébille, V, Jaber, S, Le Thuaut, A, Feuillet, F, Cinotti, R ; ATLANREA group: Extubation success prediction in a multicentric cohort of patients with severe brain injury. Anesthesiology 2017; 127:338–46 [Article] [PubMed]
Ko, R, Ramos, L, Chalela, JA . Conventional weaning parameters do not predict extubation failure in neurocritical care patients. Neurocrit Care 2009; 10:269–73 [Article] [PubMed]
Godet, T, Chabanne, R, Marin, J, Kauffmann, S, Futier, E, Pereira, B, Constantin, JM . Extubation failure in brain-injured patients: Risk factors and development of a prediction score in a preliminary prospective cohort study. Anesthesiology 2017; 126:104–14 [Article] [PubMed]
Yang, KL, Tobin, MJ . A prospective study of indexes predicting the outcome of trials of weaning from mechanical ventilation. N Engl J Med 1991; 324:1445–50 [Article] [PubMed]
Rishi, MA, Kashyap, R, Wilson, G, Schenck, L, Hocker, S . Association of extubation failure and functional outcomes in patients with acute neurologic illness. Neurocrit Care 2016; 24:217–25 [Article] [PubMed]