Prevalence, Intensity, and Clinical Impact of Dyspnea in Critically Ill Patients Receiving Invasive Ventilation

<b>Rationale:</b> Dyspnea is a traumatic experience. Only limited information is available on dyspnea in intubated critically ill patients. <b>Objectives:</b> Our objectives were <i>1</i>) to quantify the prevalence and severity of dyspnea; and <i>2</i>) to evaluate the impact of dyspnea on ICU length of stay and post-traumatic stress disorder (PTSD) 90 days after ICU discharge. <b>Methods:</b> This was a prospective cohort study in 10 ICUs in France. In patients intubated for more than 24 hours, dyspnea was quantified with a visual analog scale (from 0 to 10) as soon as they were able to communicate, the following day, and before spontaneous breathing trials. PTSD was defined by an Impact of Event Scale-Revised score of at least 22. <b>Measurements and Main Results:</b> Among the 612 patients assessed, 34% reported dyspnea, with a median dyspnea rating of 5 (interquartile range, 4-7). ICU length of stay was not significantly different between patients with versus without dyspnea (6 [3-12] and 6 [3-13] days, respectively; <i>P</i> = 0.781). Mortality was not different between groups. Of the 153 patients interviewed on Day 90, a higher proportion of individuals with probable PTSD was observed among patients who were dyspneic on enrollment (29% vs. 13%; <i>P</i> = 0.017). The density of dyspnea (number of dyspneic episodes divided by time from enrollment to extubation) was independently associated with PTSD (odds ratio, 1.07; 95% confidence interval, 1.01-1.13; <i>P</i> = 0.031). <b>Conclusions:</b> Dyspnea was frequent and intense in intubated critically ill patients. ICU length of stay was not significantly different among patients reporting dyspnea, but PTSD was more frequent at Day 90. Clinical trial registered with www.clinicaltrials.gov (NCT02336464).