Research

Changeovers of vasoactive drug infusion pumps: impact of a quality improvement program

Laurent Argaud¹ , Martin Cour¹ , Olivier Martin¹ , Marc Saint-Denis¹ , Tristan Ferry¹ , Agnes Goyatton² and Dominique Robert¹

¹Hospices Civils de Lyon, Hôpital Edouard Herriot, Department of Emergency and Medical Intensive Care, Lyon F-69003, France

²Hospices Civils de Lyon, Hôpital de la Croix-Rousse, Medical Intensive Care Unit, Lyon F-69004, France

author email corresponding author email

Critical Care 2007, 11:R133doi:10.1186/cc6209

Published:	28 December 2007

Abstract

Background

Hemodynamic instability following the changeover of vasoactive infusion pump (CVIP) is a common problem in the intensive care unit. Several empiric methods are used to achieve CVIP. We hypothesized that the variation in these procedures could generate some morbidity. We sought to assess the effects of the standardization of practice, as a quality improvement program, on the CVIP-induced incidents.

Materials and methods

We performed a prospective before-and-after intervention study including all adult patients with a diagnosis of cardiovascular failure who received a continuous infusion of vasoactive drugs or inotropic drugs. After a baseline preimplementation period (phase 1), a standardized 'quick change method' of CVIP using two syringe drivers was implemented in our intensive care unit (phase 2). Endpoints (rate and distribution of incidents: variations of systolic blood pressure >20 mmHg or heart rate >20 beats/min, and arrhythmias) were registered in both 3-month phases.

Results

We studied a total of 913 CVIP events (phase 1, 435 events; phase 2, 478 events) from 43 patients. Patient characteristics were not significantly different among phases, with a majority of the patients having septic shock. The frequency of incidents was significantly (P < 0.0001) reduced in phase 2 (5.9%, n = 28) versus phase 1 (17.8%, n = 78). This effect was observed whichever catecholamine was used. More than 98% of incidents were blood pressure variations, with a similar distribution of the nature of incidents in both phases.

Conclusion

The present study illustrates that adverse events are common following CVIP, and illustrates the positive impact of a quality improvement program to enhance inpatient safety related to this current process of care.