The authors reply
After probing the RESTORE dataset, we identified 2% of patients (49/2,449) who received antidelirium medications (n = 46) and/or experienced hallucinations or clinician-reported delirium (n = 9; three did not receive antidelirium medications), adverse events that were not prespecified. Most patients (82%) with “presumed delirium” were in our oldest age group (6–17 yr) and were cognitively appropriate for age (82%). The onset of the presumed delirium was on study day 5 (median; interquartile range, 2–12). Only approximately one-third (n = 18) were ever at-risk for IWS. Among these 18 patients, five experienced presumed delirium only after the start of weaning. These five patients were assessed for IWS using the withdrawal assessment tool (WAT)-1 (3) for 3–12 days and exhibited a variety of IWS symptoms. We have no way to retrospectively identify hypoactive delirium in the RESTORE dataset and believe we are underreporting hyperactive delirium, but most of the presumed delirium we captured did not overlap with IWS.
Interpreting clinical symptoms in the context of the patient’s trajectory of illness is crucial when differentiating IWS from delirium. Organ dysfunction, including the brain, is common in critically ill patients. In addition, shortly after PICU admission, most patients experience circadian disruption and other exposures that influence delirium risk. The primary challenges during the acute phase of illness are to manage pain and agitation while stabilizing the patient. Ensuring patient comfort remains a priority throughout the course of illness and may require upward titration of both analgesics and sedative medications as tolerance develops. The greatest potential for overlap between IWS and delirium exists during the weaning phase, when the gradual cessation of analgesic and sedative medications coupled with the stress of a protracted PICU stay may trigger one or both conditions.
Our work represents a step toward identifying risk factors for IWS that could drive further prevention efforts. The WAT-1 was developed in patients known to be weaning from opioids and/or benzodiazepines (3). Our subspecialty now has several standardized assessment tools that will allow for prospective monitoring of delirium (4–6). Going forward, we recommend that future studies of IWS in patients weaning from sedation include delirium monitoring throughout the patient’s course of illness to allow prospective study of factors associated with both phenomena. Such research should account for variation in analgesia and sedation practices, patient response to medications, the PICU environment, and the patient’s trajectory of illness. We appreciate the opportunity to address the complex interplay between IWS and delirium in the PICU, which we hope will form the basis for a robust program of research that will improve care for all critically ill children.