So, thank you, Dr. Barretto, for introducing me and allowing me to speak today, and I'm excited to present this research to you all. So, these are my co-investigators that helped tremendously with this project, so I want to give all of them a big shout out. So, just to give you all some background, the MindUSA study is a prospective, randomized, double-blind, placebo-controlled trial of adult ICU patients with acute respiratory failure or shock with delirium. Haloperidol was compared to Ziprasidone and to placebo. No difference was found in delirium or coma-free days, delirium days, coma days, 90-day survival, liberation from mechanical ventilation, ICU discharge, hospital discharge, or hospital readmission. The PADIS guidelines do not recommend to use typical or atypical antipsychotics for the treatment of delirium. Now, granted, this study came out before the PADIS guidelines, but a worldwide study of ICU practitioners revealed that typical and atypical antipsychotics were used in 65 percent and 53 percent of patients with delirium. So, QTC interval prolongation secondary to antipsychotics has not really been robustly studied in critically ill patients. So, the purpose of this study is to investigate the influence of antipsychotic medications on the change in QTC interval within critically ill delirious patients, the impact of sex on QTC interval variation, and lastly, whether QTC interval prolongation was associated with fatal ventricular arrhythmias. Our hypothesis was that antipsychotics prolong the QTC interval, but these changes are not linked with adverse outcomes or with sex. So, our methods, this was a priority with FDA oversight, a nested analysis of a multicenter randomized controlled trial that included 566 randomized ICU patients from 16 different sites who had delirium. Of these, 192 received haloperidol, 190 received zeprazidone, and 194 received placebo. So, the patients were randomized to either haloperidol, zeprazidone, or to placebo. Delirium was assessed using the confusion assessment method in the ICU or the CAM ICU twice daily, and we monitored pre-randomization 12 lead EKGs, and we had FDA approval secondary to sending in data from our first 189 patients to enroll and continue study drug in patients with a QTC up to 550 milliseconds. So, if after they got the 12 lead EKG, if greater than an hour had elapsed from that, then patients had to have their EKG verified via telemetry before they could receive study drug. For those patients that had a QTC greater than or equal to 550, then they had to be evaluated for reversible causes as to why their QTC was long, and those had to be treated before the patient was able to receive study drug. If the patient's QTC did not get less than 550 within the five days of the pre-randomization study, they were never actually randomized. So, after that the study drug was administered, then within 10 to 30 minutes afterwards, patients had QTC measured. And so, of these values that were measured, we took the highest three of these and took the mean of that, and we also looked at the highest one of these as well. And when we looked at their pre-drug administration QTC, and then we looked at their post-administration QTC, if this value had increased by more than 50 milliseconds, then we got another 12 lead EKG to verify and make sure that this was accurate. So, during the intervention phase, the medication was up titrated. So, patients could be up titrated as high of a dose to Haldol 10 milligrams twice a day, so a total of 20 milligrams a day. And they also could be up titrated to a Zeprazidone dose of 20 milligrams twice a day or 40 milligrams daily if they continued to be CAM-ICU positive. So, if they were CAM-ICU negative, they actually had their dose decreased down to a dose as low as 0.25 milligrams of either Haloperidol or Zeprazidone. And if they were CAM-ICU negative for four consecutive values, their drug was actually stopped. However, they could receive drug again if they became CAM positive again within the 14-day study period. So, the QTC interval was monitored prior to each dose of medication for 96 hours, and if the study drug was held, it was restarted at half the previous dose. And tersades, depoints, or another ventricular arrhythmia were reasons for permanent discontinuation of the study drug. So, our outcomes, our primary outcome was a median change in QTC interval between day one and day two. Secondary outcomes included association of treatment group on post-randomization day two maximum pre-dose QTC, association of sex on QTC interval on day two of treatment, the incidence of ventricular arrhythmia as a cost treatment groups, and we also looked at the correlation between the 12-lead EKGs and the telemetry EKGs as well. So, going through our statistical analysis, continuous variables were summarized as medians with interquartile ranges, and categorical variables were summarized as numbers with percentages. We did multivariate analysis were reported with odds ratios and 95 percent confidence intervals, and we collected and managed all our data using REDCap. All analysis were performed by using statistical software R. So, we did multivariate proportional odds regressions to estimate the effect of antipsychotics on next day QTC, and we justified for several pre-specified baseline covariates, including pre-randomization QTC, post-randomization max pre-dose QTC, the treatment group, age, sex, their body mass index, and their SOFA score as well. This similar model was used to determine the effects of sex on QTC. We obtained the median of the mean daily pre-dose QTC interval across all study days to describe the overall trend of pre-dose QTC interval over time, and also we conducted redundancy analysis were performed to ensure that no covariates could completely explain any of the other covariates, and the initial post-dose QTC interval was analyzed using proportional odds regression, adjusting for the same pre-specified covariates that I told you about before. So, lastly, a Spearman correlation was used to determine the correlation between the bedside telemetry QTC and the EKG 12-lead measure QTC when study drug was held due to QTC interval prolongation. So, our results are as follows. At baseline, our median QTC intervals across the haloperidol, zeprasin, and placebo treatment groups were as follows, 458, 451, and 452, respectively. The median change from day one to day two were negative one, zero, and negative three and a half, respectively, by treatment groups, and the median days of exposure to an antipsychotic study drug or placebo was four. So, compared with placebo, neither haloperidol nor zeprasidone significantly affected the next day QTC interval, and this effect was not significantly affected by sex either. So, TRSADs occurred twice, and both were in the haloperidol group, but it's important to note that these were both more than four days post-exposure, and given the half-life of haloperidol, I think we can assume that these were unrelated to our study drug. So, now, moving on with the results, as you can see here, baseline QTC is really what affected next day QTC. Treatment drug, age and enrollment, sex, BMI and SOFA, and also pre-randomization baseline QTC did not. There was no change in QTC by early exposure to these drugs. So, if you look at the y-axis, this is day two maximum pre-dose QTC interval, and if you look at the placebo, haloperidol, and zeprasidone, there's really no change in QTC There's also no change in QTC interval by longitudinal exposure to drugs either. So, if you look at the x-axis here, this shows you day one through day 15 of the study drug exposure, and there's really no change in the QTC interval. There is a fairly good association between EKG and telemetry QTC interval, as you can see here. So, in summary, baseline QTC affects next day QTC. QTC is not affected by early or by longitudinal exposure to antipsychotics, and EKG and telemetry QTC correlate fairly well. So, strengths of this study include that this, to our knowledge, is really the largest analysis of antipsychotic effects on QTC intervals in critically ill patients. It's really externally generalizable, considering that this was conducted at 16 different centers. The frequency of QTC interval monitoring was rigorous, with telemetry measurements validated using a 12-lead EKG, and critically ill patients have many different electrolyte abnormalities and many different ICU drug confounders, making them a very complicated patient population. Our large cohort allowed for investigation of sex on QTC interval. Also, as many of you all know, the QTC interval can have diurnal variation. You do expect it to be highest in the morning, and since our study drug was given twice a day, we also were able to account for that. So, limitations of this study included that ICU drug confounders and other electrolyte abnormalities were not collected. However, these were critically ill patients, so I think that most of us can expect that many of these patients were on QT prolongers, such as amiodarone or maybe vureconazole. We used the Bizette correction from telemetry to provide a heart rate-corrected QTC, and maybe the Fridericia or the Framingham would be better to use in a patient that had faster heart rates. There is a possibility that some of these patients had long QT syndrome, and we did not do genetic testing. However, looking at our inclusion criteria, you would not have expected these patients to met enrollment criteria for this study. And then, lastly, there can be some EKG variability. Newer EKGs use multi-lead acquisition versus older ones use single-lead acquisition, which could lead to some variability. So, in conclusion, antipsychotics did not increase QTC intervals beyond that of placebo. Ventricular arrhythmias were not associated with antipsychotic administration. Telemetry and EKG measured QTC intervals correlate fairly well, and the frequency of current QTC monitored due to antipsychotic administration may likely be excessive. So, this concludes my presentation. I'm happy to take any questions that anyone has at this time.

antipsychotic medications

QTC interval change

critically ill patients

delirium

QTC monitoring