The association between nurse-administered midazolam following cardiac surgery and incident delirium: An observational study

Priscilla G Taipale, Pamela A Ratner, Paul M Galdas, Carol Jillings, Deborah Manning, Connie Fernandes, Jaime Gallaher

Research output: Contribution to journalArticlepeer-review


BACKGROUND: Post-operative delirium after cardiac surgery is an adverse event that affects patients' recovery and complicates the delivery of nursing care. Numerous risk factors for delirium are uncontrollable; however, nurses' pro re nata drug administration of sedatives may be a controllable risk factor. OBJECTIVES: This study examined the relationship between nurses' pro re nata administration of midazolam hydrochloride to cardiac surgery patients and the development of post-operative delirium. DESIGN: Observational study. SETTING: Cardiac surgery intensive care and nursing units of a tertiary care center in Vancouver, Canada. Participants: 122 male and female patients requiring non-emergent surgery for coronary artery disease or valvular heart disease who did not have pre-existing cognitive impairment, severe hearing or visual impairment, substance misuse, alcohol intake exceeding 7 drinks per week, or renal impairment requiring hemodialysis. METHODS: Patients were assessed for delirium, on three occasions, with the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) during the first 72h after surgery and through reviews of physicians' notes. Risk factor and midazolam dosage data were collected from medical records. RESULTS: 77.9% of the patients in this sample received midazolam hydrochloride post-operatively. The prevalence of delirium ranged from 37.7% to 44.3%. Almost all of the dosages of midazolam (85-87%) were given before the first indication of delirium; that is, most of the patients had received their entire dosage before the first signs of delirium were detected. Bivariate analysis with logistic regression models revealed that for every additional milligram of midazolam administered, the patients were 7-8% more likely to develop delirium. Multivariate logistic regression models demonstrated that the magnitude of the association between midazolam dosage and delirium was not confounded by established risk factors including age and peripheral vascular disease. CONCLUSION: Nurses play an important role in the prediction, assessment and prevention of post-operative delirium. Sedatives should be administered with caution because they increase a patient's risk of developing delirium. Nurses' decisions regarding sedation administration must be informed by empirical knowledge, accurate assessment data and clear rationale with consideration of how these actions may contribute to the development of delirium.
Original languageEnglish
Pages (from-to)1064-1073
Number of pages10
JournalInternational Journal of Nursing Studies
Issue number9
Publication statusPublished - Sept 2012

Cite this