Agreement Between ICU Clinicians and Electrophysiology Cardiologists on the Decision to Initiate a QTc-interval Prolonging Medication in Critically Ill Patients with Potential Risk Factors for Torsade de Pointes: A Comparative, Case-Based Evaluation


  • This study was supported by departmental funds.

For questions or comments, contact John W. Devlin, School of Pharmacy, Northeastern University, 360 Huntington Avenue, R218 TF, Boston, MA 02115; e-mail:


Study Objectives

To measure concordance between different intensive care unit (ICU) clinicians and a consensus group of electrophysiology (EP) cardiologists for use of a common rate-corrected QT interval (QTc)-prolonging medication in cases containing different potential risk factor(s) for torsade de pointes (TdP).


Prospective case-based evaluation.


Academic medical center with 320 beds.


Medical house staff (MDs) and ICU nurses (RNs) from one center and select critical care pharmacists (PHs).


Completion of a survey containing 10 hypothetical ICU cases in which patients had agitated delirium for which a psychiatrist recommended intravenous haloperidol 5 mg every 6 hours. Each case contained different potential risk factor(s) for TdP in specific combinations. A group of five EP cardiologists agreed that haloperidol use was safe in five cases and not safe in five cases.

Measurements and Main Results

For each case, participants were asked to document whether they would administer haloperidol, to provide a rationale for their decision, and to state their level of confidence in that decision. Most clinicians (92 of 115 [80%]) invited to participate completed the cases. Among the five cases where EP cardiologists agreed that haloperidol was not safe, 29% of respondents felt that haloperidol was safe. Conversely, in the five cases where EP cardiologists felt haloperidol was safe, 21% of respondents believed that it was not safe. Overall respondent-EP cardiologist agreement for haloperidol use across the 10 cases was moderate (κ = 0.51). MDs and PHs were in agreement with the EP cardiologists more than RNs (p=0.03). Interprofessional variability existed for the TdP risk factors each best identified. Clinician confidence correlated with EP cardiologist concordance for MDs (p=0.002) and PHs (p=0.0002), but not for RNs (p=0.69).


When evaluating use of a QTc interval-prolonging medication, ICU clinicians often fail to identify the TdP risk factors that EP cardiologists feel should prevent its use. Clinician-EP cardiologist concordance varies by the specific risk factor(s) for TdP and the ICU professional conducting the assessment.