METROLOGY for DRUG DELIVERY

GOOD PRACTICE GUIDE:

Calibration of Medical Infusion Pumps

www.drugmetrology.com
This good practice guide was written as part of activity A5.3.1 from the EMPIR Metrology for Drug Delivery
(MeDD II) project. The three-year European project commenced on 1st June 2019 and focused on providing
traceable measurements of volume, flow and pressure of existing drug delivery devices and mixing behaviour
and occlusion phenomena in multi-infusion systems. For more details about this project, please visit
www.drugmetrology.com

This good practice guide was produced by:

Dr. Emmelyn Graham TÜV SÜD National Engineering Emmelyn.graham@tuvsud.com

Laboratory (NEL), UK
Dr. Zoe Metaxiotou NQIS / EIM, Greece zoe@eim.gr
Elsa Batista IPQ, Portugal ebatista@ipq.pt
Contents

Why is Calibration Important? ......................................................................................................................... 1

How to Calibrate Infusion Pumps..................................................................................................................... 1
Common Issues with the Calibration of Infusion Pumps ................................................................................ 2
Performance Characterisation of Infusion Pumps ........................................................................................... 3
Multi-Infusion Setups ....................................................................................................................................... 4
New Calibration Methods for Ultra-Low Flow Rates....................................................................................... 4
Summary ........................................................................................................................................................... 5
Further Reading ................................................................................................................................................ 5
References......................................................................................................................................................... 5
Why is Calibration Important? given to neonates, which normally occurs at
There have been numerous injuries, deaths and very low flow rates. Although the calibration
adverse health effects associated with the use of principles highlighted in this guide refer to
infusion pumps which have highlighted syringe pumps, they can be applied to other
significant safety issues [1]. For example, types of pump.
approximately 80-90 % of hospitalised patients
in the UK receive intravenous (IV) therapy using
infusion devices to deliver medication, fluids and
nutrients into patients [2,3]. It is extremely
important that the delivery of medication and
other fluids is precisely controlled over time, and
the delivered dose is accurately known,
especially for critical drugs at high
concentration. Regular calibration and
maintenance of infusion pumps enables the
identification of any issues with equipment and
ensures correct dosage to patients, minimising Syringe pump for medical use
potential safety risks.
How to Calibrate Infusion Pumps
A common calibration method to determine
the flow rate error of an infusion pump involves
the use of an infusion device pump analyser,
which can give information on the flow rate,
volume and pressure. The analyser functions as
a master calibrator to quickly test infusion
pumps performance, however, it is important
that it is calibrated regularly. This method is
used in hospital facilities.

The gravimetric method is used extensively in

the laboratory by National Metrology Institutes
(NMIs) and accredited laboratories, as a very
Syringe infusion pumps and incubator for new- accurate way to calibrate pumps and flow
borns in hospital meters. This method uses a balance to weigh
Typical infusion pumps include: the mass of liquid (i.e. water, the specification
of water is defined in ISO 3696 [4]) that is
delivered by the pump into a weighting vessel
 Syringe
on the balance. As the density of the water is
 Peristaltic pump delivery
known at the temperature of the test (usually
 Gravity controlled delivery
20 °C), this is used to calculate the volume of
 Insulin
liquid delivered (volume = mass/density). The
 Patient controlled analgesia (PCA)
volumetric flow rate is determined from the
 Elastomeric piping
quotient of the total liquid volume and time
 Pain killing drugs
taken for the delivery of that liquid. Details on
 Enteric nutrition
how to test infusion pumps are provided in IEC
60601-2-24:201 [5]. This standard provides
Although there are many different types of
criteria for testing the performance and safety
infusion pump, this guidance document focuses
of pumps including detailed calculations on
on critical drug delivery applications which
how to assess the performance using a
require a high degree of accuracy. Such
gravimetric calibration method. However, it
applications include drugs with a narrow
should be noted that this standard is currently
therapeutic margin, drugs with a short half-life,
under review and, therefore, subject to change.
high concentration drugs and any infusions
1
A simplified schematic of the gravimetric can be used to reduce evaporation [8]. During
calibration setup is shown below. the calibration, environmental effects on the
balance performance, such as draughts and
Syringe
temperature fluctuations, should be
considered.
Syringe pump
g 0.0000 g Balance The relative flow rate error of a pump is given
by:

! #$% & '% − ($ $($')$

= ∗ 100%
From drug metrology research conducted by
several NMIs, it is recommended that the ($ $($')$
equation provided in Appendix B of ISO 4787
[6], the international standard which gives
detailed information on how to accurately
Where Qpump set point is the target flow rate set on
calibrate volumetric equipment, is used in
order to more accurately determine the the pump and Qreference is the actual flow rate
volume of the liquid mass on the balance. The determined in the calibration. A negative error
equation in ISO 4787 includes additional factors means the pump is delivering over its target
that take into account the effect of air flow rate and a positive error means that it is
buoyancy and the thermal expansion of the delivering less fluid than desired. This
disposable syringe [6,7]. The volumetric flow definition of error complies with metrological
rate is determined from the quotient of the standards [9, 10].
total liquid volume and the time taken for the
delivery of the liquid. However, it should be noted that a different
definition is used in medical standards, such as
The volumetric flow rate (Q) can be expressed in IEC 60601-2-24:201 [5], where the error is
as: given by:

1 1 ($ $($')$ − ! #$% & '%

= − × × 1− × 1− − 20 = ∗ 100%
− − ! #$% & '%

Where tf is the final time, ti is the initial time, If

is the final balance reading, Ii is the initial
It is important to identify how the error is
balance reading, ρw is the density of water used
calculated as otherwise false corrections may
for the calibration, ρA is the density of air, ρB is
the density of the mass standards used to be applied. More detailed information on these
calibrate the balance, ϒ is the coefficient of conflicting definitions is provided in [11].
thermal expansion of the disposable syringe
and T is the temperature of the water used.
Common Issues with the Calibration
More information on the expansion coefficients of Infusion Pumps
is provided in ISO 4787.  Air bubbles – a sufficient volume of water
should be passed through the piping to
It is recommended that additional correction purge any air bubbles.
factors should be used including a correction  Flow stability and pump response time –
for the evaporation of water during the there is a need to wait until the target
calibration and a correction for the buoyancy infusion flow rate has been reached, as
effect of the needle immersed in the water in large errors will occur before this. The
the beaker. To reduce the evaporation of the time taken before the required target flow
water during the calibration, the use of special rate is achieved is often referred to as the
evaporation traps, similar to those used for start-up delay or delayed onset and
piston pipette calibrations, can be used. depends on the flow rate, type of pump
Alternatively, a layer of oil, such as paraffin, on and disposables used. More information
top of the water, as described in ISO 7886-2, on this can be found In IEC 60601-2-
2
 24:201. These factors are illustrated in the permissible error. The uncertainty of the
diagram below: calibration must also be considered in
metrological conformity. The uncertainty of
the measured value (e.g. 10 mL/h), could be ±1
Start-up delay
% at the 95 % confidence level. This means the
Flow Rate

true reference flow rate can be anywhere

within 10.1 and 9.9 mL/h. Hence, the maximum
Target flow rate
Actual flow permissible error is a combination of the
measured error and the uncertainty. ISO 7886-
Time 2 specifies a long-term accuracy of 2 % for the
 Incorrect syringe diameter - this can lead percentage flow rate error when measured
to large dosing errors, e.g. 5 % error in the over a one-hour time period, a short-term
syringe diameter can lead to a 10 % error accuracy of 5 % for the maximum flow rate
in the delivered drug flow rate. Different variation when measured over a two-minute
syringe models and brands can have time period, and an accuracy of 2 % when
different diameters, so it is always measured over a five-minute time period [8].
important to check this.
 Water evaporation – if not controlled This highlights the need for a low measurement
properly this can cause an uncertainty on calibration techniques to assess
underestimation of flow rate the metrological performance of pumps.
measurements. Uncertainty contributions in the calibration of
 Fluid connection leaks - this refers to the pumps are attributed to:
incorrect connection between the tubes,
the disposables and the pump and will  Environmental conditions changing,
result in a reduced measurement of flow e.g. room temperature, drafts,
rate. humidity etc.
 Equipment compliance – this refers to  Repeatability of measurements
expansion effects in the components from (spread in repeated measurement
the elasticity of the materials, e.g. the data)
expansion of a plastic syringe when the  Flow rate instability
pressure increases during the start-up of
 Resolution of calibration equipment
the pump. Expansion of components can
used
also occur when the flow rate is increased;
this increase in the pressure can cause the  Calibration of the measurement
infusion line, syringe, and other equipment (e.g. balance, temperature
equipment to expand. The expansion sensors)
results in the internal volume increasing  Equations used to determine density
hence can cause a delayed response for and expansibility
flow rate changes. ISO 7886-2 [8] provides  Buoyancy corrections for the needle
information on measuring the compliance submerged in the collected liquid
of a syringe; however, it does not take into  Air buoyancy corrections
account additional contributions such as  Leakage through the tubing or
the infusion lines, which can also expand. connectors
 Drift of equipment
Performance Characterisation of  Timing errors
Infusion Pumps  Evaporation corrections
The assessment of the performance of a pump  Calibration technician experience
being calibrated cannot be entirely based on  Auxiliary equipment used and the
the determination of the error even if it is compliance (elastic expansion) effect
found to be within ±2 %, which is normally (e.g. infusion lines)
stated by the manufacturer as the maximum

3
  Dead volume – this is the volume of of drugs assessed for different setups [12].
liquid contained between the pump Advanced modelling and validation are
under test and the reference currently being investigated by NMIs as part of
measurement device. If the a collaborative European research project on
temperature of the connecting tubing drug delivery metrology.
and fluid are different at the start of a
calibration from the corresponding
temperature at the end of the
calibration, the quantity of fluid held
within this volume will have changed,
hence leading to underestimation or
overestimation of the quantity of fluid
passed between the pump and the
reference.

For the calibration of a single pump in the

laboratory, there will be very low back pressure
as the liquid flows into a beaker at atmospheric Patient with monitor and infusion pumps in an
pressure. However, in infusion setups, the intensive care unit
back pressure will be higher as the drugs are
flowing into veins or arteries. New Calibration Methods for Ultra-
Low Flow Rates
Multi-Infusion Setups The gravimetric calibration method is limited
Calibration techniques are used to test single when trying to measure ultra-low flow rates, as
pumps, but things get much more complicated the measurement uncertainty of this method
when using multiple pump infusions where the increases substantially and evaporation
fluids from each pump are combined before a becomes a critical error factor. Special setups
single injection point into the patient. Multi- for gravimetric calibrations have been
infusion setups are very common in hospitals performed by NMIs down to 6 µL/h (100
where several different drugs need to be nL/min) [13]. However, new developments in
administered at different dosing rates using pumping technology, e.g. insulin pumps, can
separate infusion pumps for each drug. deliver extremely low flow rates typically 10
Dosing accuracy can be affected by the µL/h and much lower. To test these new
interaction between the pumps and the dead devices, novel calibration techniques are being
volume [12]. The dead volume effect results developed by NMIs that are capable of
from the drug injected from one pump needing measuring flow rates down to 5 nL/min. Some
to travel through the infusion line between the of these new calibration techniques are based
mixing point and the inlet to the patient; this on optical measurements, for example, tracking
volume is known as the dead volume. The the displacement of a meniscus along a
interaction between the pumps and the dead capillary.
volume can cause dosing issues and preferably
this volume should be a small as possible.
Positioning the pumps closer to the patient
would be a solution but often it is necessary to
provide some flexibility for patient movement
by using longer drug delivery lines.

The effect of multi-infusion setups has been

modelled and the flow rate and concentration
4
 essential performance of infusion pumps
and controllers, 2019
7. ISO 4787:2010, Laboratory glassware —
Volumetric instruments — Methods for
testing of capacity and for use, 2010
8. E. Batista, J. A. Sousa, R. F. Martins,
Calibration of Insulin Pumps, Journal of
Diabetes and Treatment, Vol 4, Issue 03,
2019
9. ISO 7886-2:1996 Sterile hypodermic
syringes for single use — Part 2: Syringes
Insulin pump for use with power-driven syringe pumps,
1996
Summary 10. JCGM 200:2012, International vocabulary
This guidance note highlights the importance of of metrology – Basic and general
the regular calibration of medical infusion concepts and associated terms (VIM), 3rd
pumps to reduce safety risks by identifying any
edition, 2012
performance issues and ensuring the correct
11. JCGM 100:2008, Evaluation of
dose to patients. The gravimetric calibration
method is recommended as an accurate measurement data - Guide to the
method that can be used for assessing the expression of uncertainty in
performance of an infusion pump analyser and measurement (GUM), 2008
other pumps. Recommendations on how to 12. MeDDII report, Measurement error: two
perform a gravimetric calibration, referencing opposite definitions in metrology
to appropriate standards, have been outlined. standards and medical standards, 4/2021
https://drugmetrology.com/wp-
Further Reading content/uploads/2021/04/Measurement-
EMPIR Metrology for Drug Delivery website – error-defenitions_FINAL_04_2021.docx
http://www.drugmetrology.com/ 13. R. A. Snijder, M. K. Konings, P. Lucas, T.
C. Egberts and A. D. Timmerman, Flow
References variability and its physical causes in
infusion technology: a systematic review
1. US Food and Drug Administration website
of in vitro measurement and modelling
accessed 26th March 2020
studies, Biomedical Engineering-
2. P. Lucas, R. A. Snijder, A. M D. E.
Biomedizinische Technik, 60(4): 277–300,
Timmerman, E. Batista, H. Bissig, F.
2015
Ogheard, Best Practice Guide, EMPIR
14. E. Batista, A. Furtado, J. Pereira, M.
Metrology for Drug Delivery, Version 13-
Ferreira, H. Bissig, E. Graham, A.
05-2015, 2015
Niemann, A. Timmerman, J. Alves e Sousa,
3. Rigel Medical (UK), An introduction to
F. Ogheard, A. W. Boudaoud, New EMPIR
infusion pump testing, Version 1.0 - 2014
project – Metrology for Drug Delivery,
4. https://www.fda.gov/medical-
Flow Measurement and Instrumentation,
devices/general-hospital-devices-and-
72, 101716, 2020
supplies/infusion-pumps
5. ISO 3696:1987, Water for analytical
laboratory use – specification and test
methods, 1987
6. IEC 60601-2-24:2019, Medical electrical
equipment, Part 2-24: Particular
requirements for the basic safety and
5