2004 Annual Scientific Meeting - CLICK FOR PROGRAMME

Investigation of the nature and extent of flow related problems within multi-pump infusions

Professor Colin, School of Engineering and Design, Brunel University, Uxbridge, Middlesex, UB8 3PH

Introduction

The need for this work was identified by the MHRA Infusion Pump Evaluation Centre (University of Bath) following reports of unexpected refluxes of liquid(s) during medical infusions.  The project area was broadened to include an evaluation of the nature and extent of the disruption of flows of different therapeutic agents delivered simultaneously by multi-pump infusion. 

Objectives

(i)  Review current clinical practice in order to identify and assess flow related problems in infusion therapy.

(ii) Review with equipment manufacturers the problems identified by practitioners in order to establish more clearly the underlying causal factors. 

(iii) Formulate a substantive follow-on project if appropriate.

Methods

Clinical practice was reviewed by surveys which included:  telephone/face-to-face interviews with 14 practitioners;  questionnaires distributed through SIVA and ICNARC to a total of 470 physicians and 180 nurses, all known to be infusion therapy practitioners.  Interviews were also held with clinical engineers, medical professional bodies, regulatory agencies and equipment manufacturers.

Results

The questionnaire survey produced responses from 113 physicians/anaesthetists and 84 nurses.  Of these 197, 40% indicated no flow related problems, 17% identified pump only problems, and 43% reported flow related problems.  Two broad categories of flow problems were identified:

(i)   arising from operational factors: including pressure effects, wrongly routed flows, interrupted flow, blocked flow, bolus flow;

(ii)  arising from equipment problems:  e.g. cracked components (syringes, connectors), leaking seals, faulty valves.

Discussion

The ideal multi-infusion system would provide, on start-up:  instantaneous delivery of agents into the patient’s vascular system at the selected rates which would be held constant and would occur for the set time windows.  A key finding was that the principle undesirable flow effects (which cause actual or potential patient harm) arise from the complexity of dynamic interaction within multi-pump infusions.  These include:  long delays in start-up of drug in flow to the patient, non-constant flow rates, reflux of drugs/blood and unintended bolus delivery of agents.  Such features arise from the very low flow rates, significant system compliance coupled with the flow resistances and finite mixing volumes.  Multi-infusion systems are neither fully understood nor characterised, hence some aspects of infusion flow behaviour remain unpredictable with associated risks to patients.

Conclusions 

Findings from this survey project have led to the proposal of a new, substantive research project which would:

(i) through experimental investigation of a small number of selected model infusion systems:  reveal and allow characterisation of the dynamic behaviour of flows within these well defined systems;