The steady-state flow rates were normally distributed; thus, t tests were used to analyze steady-state flow rate data. The time intervals for steady-state flow rates ranged from 4 to 18 minutes. Overall, the steady-state flow rate for each trial in both groups was below the pump programmed flow rate. Figure 4 depicts the mean percent deviation from the programmed flow. This deviation was below 0 for each programmed flow rate (maximum deviation = −17.3% ± 5.8%), indicating that all flow rates were lower than the corresponding programmed flow rate. Table 3 depicts the mean steady-state flow rates for each trial. For the highest initial flow rate (1.0 mL/h), the steady-state flow rate attained was lower in the filter group than the control group for the initial rate (P = 0.04) and doubling of initial rate (P = 0.04) with a trend during the return to initial rate (P = 0.06). This effect was not observed when doubling the initial rate trials of 0.8 or 0.6 mL/h compared with the control group. There were no statistically significant differences in the steady-state flow rate between either group for any of the other flow rates (0.4, 0.6, or 0.8 mL/h), although the width for the 95% confidence intervals for these differences ranged from 0.08 to 0.73.
Flow Rate Variability
When the variance (the square of the SD) for the steady-state flow rate was compared between the filter and control groups, the only difference in rate variability (Table 1) was observed for the doubling of the flow rate to 1.2 mL/h (after initial flow rate of 0.6 mL/h) for that rate trial. This difference was not observed in any other trial (Table 3).
Time to Reach Steady-State Flow
Student t tests were used to analyze these data. The average time to reach steady-state flow from start of the infusion ranged from 1.3 to 3.8 minutes (Table 4) and did not differ statistically between the control and filter groups. In addition, the average time to reach steady-state flow after a flow rate change ranged from 0.8 to 5.5 minutes and also did not differ between the control and filter groups. The 95% confidence intervals for these differences were wide between both groups (see “Study Limitations”).
This study found that an in-line IV filter did not statistically significantly increase the amount of time from zero flow to initial flow (startup delay) and did not statistically significantly increase the amount of time to reach steady-state flow. Also, in-line IV filters did not have an effect on flow variability.
Our study revealed that, even with adherence to the syringe pump manufacturer’s recommendations, the steady-state flow rates for all experimental trials were lower than the syringe pump programmed flow rates (Table 3; Figure 4). Only for the initial flow rate of 1.0 mL/h were the steady-state flow rates lower in the filter group compared with the controls with statistical significance only present with the initial flow rate and doubling of initial flow rate and not maintained with the return to initial rate (initial rate, P = 0.04; doubling of initial rate, P = 0.04; and return to initial rate, P = 0.06). This same effect was not observed at the initial rate trials of 0.8 or 0.6 mL/h when doubling to rates >1.0 mL/h compared with the control groups. Also, there were no differences in the experimental trials with initial flow rates of 0.4, 0.6, and 0.8 mL/h. Therefore, the aforementioned observed effect may not be a sustained finding with further investigation.
Our results demonstrate that in-line IV filters did not affect startup delay under our study conditions with a mean startup delay of <1 minute across all trials. The preparatory trials (Appendix 1) determined that 2 mL of fluid delivered immediately before starting any experimental trials minimized startup delay because of mechanical factors of the syringe pump. All experimental trials were performed immediately after this 2 mL of fluid delivery from the syringe pump. Bartels et al.2 tested the same syringe pump as our study (Medfusion 3500) and illustrated the importance of immediately starting the syringe pump infusion after pump priming; their study revealed that a longer time interval between pump priming and the subsequent start of the infusion created a longer startup delay. The backflow observed during startup delay was also noted by Schmidt et al.1 In that study, the authors described transient decreases in weight measurements using gravimetric methods when a 50-mL syringe was used with a syringe infusion pump.
Time to Reach Steady-State Flow
In our study, the use of in-line IV filters did not significantly affect the time to reach steady-state flow with our time measurements ranging from 1.3 to 3.8 minutes, although the confidence limits for differences between filter and control groups were wide. After a rate change, the time to reach steady-state flow ranged from 0.8 to 5.5 minutes at all flow rates. Neff et al.3 found that infusate delivery was significantly influenced by the syringe size and the programmed flow rate; lower pump programmed flow rates and larger syringe sizes used for delivery from the syringe pump significantly prolonged the time to reach steady-state flow. In another study, the programmed flow rate of 1.0 mL/h was not achieved before 60 minutes when a 50-mL syringe connected to the syringe pump was used for delivery.1
Steady-State Flow Rates Compared with Pump Programmed Flow Rates
Delivered steady-state flow rates were lower compared with programmed flow rates whether a filter was present or not. Although syringe infusion pumps represent the most accurate device for delivery at low flow rates, pump performance may be compromised4,5 when operational limits are approached (eg, with very low flow rates). Some of these operational limitations involve the pumps’ components and can lead to startup delays and a longer time to reach steady-state flow.1,3,6–10
To evaluate the effect of filters on syringe pump performance, Jonckers et al.11 compared the effect of in-line IV filters (0.2-μm size) with controls without a filter on the in-line pressures. When in-line pressures exceed a preset limit, this would cause the syringe pump to alert the user to a possible occluded syringe pump delivery system. A filter added to the system was associated with higher in-line pressures11 that Jonckers’ group hypothesized was because of increased resistance from the clogging of filter pores. Under our study conditions, steady-state flow rates were maintained after the addition of a filter, although at the highest flow rate of 1.0 mL/h, the steady-state flow rate attained was lower in the filter than the control group for the initial rate (P = 0.04) and doubling of initial rate (P = 0.04) with a trend during the return to initial rate (P = 0.06). Because this effect was not observed when doubling the initial rate trials of 0.8 or 0.6 mL/h compared with the control group, it brings into question the validity of these findings.
Flow Rate Variability
Flow rate variability was similar in the filter and control groups in all conditions except for the doubling of the flow rate to 1.2 mL/h (after initial flow rate of 0.6 mL/h), which showed decreased variability (Table 3). These results were not sustained across the entire trial and raise questions about their reliability.
Transient and random flow rate alterations were observed during some of the trials (Figure 3, A and C). Each measurement was graphically displayed in real time after input into the database, allowing the study investigators to search for contributing factors present during the experimental trials. No factors were noted that could explain the rate alterations seen (eg, sudden changes in laboratory environment). Further investigation would be required to examine other internal or external factors that might influence the syringe infusion pump system delivery profile. These yet-to-be-determined factors may be present in patient care areas that use syringe infusion pumps.
Clinical Impact of IV In-Line Filters
Filters may lead to decreased complication rates, lengths of stay, and organ dysfunction in pediatric intensive care unit patients12,13 and may significantly reduce overall complication rates in sick newborns.14 A 2015 Cochrane analysis involving 704 preterm infants and neonates15 revealed a nonsignificant reduction in individual complications (eg, sepsis) and may have been underpowered to detect low incidence complications. The 2015 Cochrane analysis15 concluded that there was “insufficient evidence to recommend the use of IV in-line filters to prevent mortality and morbidity in neonates.”
The American Society for Parenteral and Enteral Nutrition, the British Pharmaceutical Nutrition Group Working Party, and the U.S. Food and Drug Administration recommend IV filter use for parenteral nutrition to avoid the potential hazards of precipitation, particulate matter, air, and micro-organisms.16,17 The 0.2-μm size filter is recommended for nonlipid-containing admixtures, whereas the 1.2-μm size filter is recommended for lipid-containing solutions.
In our study, the following factors were controlled to minimize confounding variables: (1) a 10-mL syringe was used for all experimental trials to minimize effects related to syringe size differences; (2) the IV infusion system was standardized; (3) new IV infusion system parts and filters were used for each experimental trial with the exception of the single-lumen central line; (4) assembly and priming methods of the IV infusion system, and all experimental techniques, were standardized; (5) the syringe pump was positioned at the same height as the distal tip of the single-lumen catheter; and (6) environmental conditions were consistently maintained (eg, evaporation was negligible). Limitations of our study include analyzing only 1 syringe infusion pump brand and 1 size (0.2 μm) filter. In addition, this study may have been underpowered to detect smaller differences in time to reach steady state.
Syringe Pump Infusion System Performance
Strategies to optimize pump performance have been described.7 Improving recognition of the limitations, safety issues, and other factors that impact syringe pump performances are a key strategy in the U.S. Food and Drug Administration’s Infusion Pump Improvement Initiative.c Syringe infusion pumps and in-line IV filters are 2 parts of the continuous IV drug delivery system for complex surgical and critically ill pediatric patients. Infusion system dynamics are complex with the potential for greater clinical impact in these very ill and low-body-weight pediatric patients. For instance, the very low body weight of the patient may result in limited IV access and restricted total daily fluid. Lovich et al.18–25 have extensively investigated drug delivery kinetics and the dynamics of continuous infusion therapy. Relevant factors influencing continuous drug delivery include the specific architecture of the connectors, the manifold design for multiple infusions, the selection of the infusion connecting port, the dead volume of the system, the presence of a carrier fluid, alterations in the infusate and carrier flow rates, and coinfusion interactions.18–25 Health care providers should be educated regarding syringe pump devices and limitations within their own patient care areas to enable the best clinical decisions for delivery of medications to complex surgical and critically ill pediatric patients.
The results of this study demonstrate that IV filters do not have statistically significant effects on syringe pump performance in the delivery of infusate at the low flow rates that are commonly used in the pediatric intensive care unit or in the operating room for a complex pediatric surgical patient as it relates to startup delay, time to steady-state flow, and flow rate variability. The overall flow rate was lower than the programmed flow rate with or without a filter.
Before full implementation of the experimental trial protocol involving the control and the filter groups, a series of separate preparatory trials were performed to establish the steps required to assure optimal syringe pump performance under reproducible conditions. These preparatory trials were designed to minimize the startup delay component attributable to the mechanical properties of the syringe pump (eg, engaging of the gears in the mechanical drive of the specific syringe pumps) used in this study. Filters were not used during any of these preparatory trials. Startup delay was defined as the time after initiation of the infusion from zero flow to initial fluid flow at the distal tip of the catheter (see the graphical representation in Figure A1). By using a flow rate of 1.0 mL/h, 6 preparatory trials were performed. Three of these experimental trials, “trials without pump priming,” involved manually priming the IV infusion system with NS and the syringe pump only used for infusion delivery at 1.0 mL/h per preparatory trial protocol. In the other 3 experimental trials, “trials with pump priming,” the IV infusion system was manually primed with NS followed by the additional step of “pump priming” before starting the infusion delivery at 1.0 mL/h. Pump priming followed the manufacturer’s instructions of using the syringe pump at a preset rate of 300 mL/h while observing for fluid movement at the distal end of the single-lumen catheter. A pump priming volume of 2 mL was used for each experimental trial. For all 6 experimental trials, the infusion was started at a programmed flow rate of 1.0 mL/h until steady-state flow was reached.
Weight measurements were obtained at the start of the 1.0-mL/h infusion flow rate and then every 2 minutes thereafter. For each time interval, the increase in the measured weight was converted to volume by multiplying by 1 g/mL (the density of NS at 22°C is 1.0046 g/mL, which was rounded to 1 g/mL). The flow rate was then calculated by dividing the volume by the time for each interval. These flow rates were graphed against the time elapsed in real time.
Data were collected every 2 minutes with a graph generated in real time that plotted the delivered flow rate of fluid per unit time. Steady-state flow was defined as being achieved when 4 consecutive graphical data points 2 minutes apart stabilized around a flow rate value with at least 1 data point above and 1 below this value. The first of these 4 data points was used to identify the start of the steady-state flow (Figure A1).
We examined differences between preparatory experimental trials comparing trials “without pump priming” and trials “with pump priming” (Figure A2). Startup delay was significantly greater in the trials “without pump priming” (4.7 ± 1.2 minutes) compared with the trials “with pump priming” (0 ± 0 minute; P = 0.02). The preparatory trials “with pump priming” did not have any startup delay (0 ± 0 min), whereas every trial “without pump priming” had a startup delay with concomitant backflow (mean backflow = −0.11 g/mL ± 0.05).
The mean time to steady-state flow was lower in the group “with pump priming” (3.3 ± 1.2 minutes), achieving steady-state flow within 5 minutes after the start of the infusion. In contrast, the trials “without pump priming” had longer mean times to steady-state flow (12.7 ± 3.1 minutes; P = 0.02), requiring up to 15 minutes to achieve steady-state flow.
Name: Destiny F. Chau, MD.
Contribution: This author helped design the study, conduct the study, and write the manuscript.
Name: Terrie Vasilopoulos, PhD.
Contribution: This author helped statistical analysis, comparison of the data, and write the manuscript.
Name: Miriam Schoepf, MD.
Contribution: This author helped to analyze the data and write the manuscript.
Name: Christina Zhang.
Contribution: This author helped conduct the study and write the manuscript.
Name: Brenda G. Fahy, MD, MCCM.
Contribution: This author helped design the study, analyze the data, and write the manuscript.
This manuscript was handled by: James A. DiNardo, MD.
1. Schmidt N, Saez C, Seri I, Maturana A. Impact of syringe size on the performance of infusion pumps at low flow rates. Pediatr Crit Care Med 2010;11:2826.
2. Bartels K, Moss DR, Peterfreund RA. An analysis of drug delivery dynamics via a pediatric central venous infusion system: quantification of delays in achieving intended doses. Anesth Analg 2009;109:115661.
3. Neff SB, Neff TA, Gerber S, Weiss MM. Flow rate, syringe size and architecture are critical to start-up performance of syringe pumps. Eur J Anaesthesiol 2007;24:6028.
4. Capes DF, Dunster KR, Sunderland VB, McMillan D, Colditz PB, McDonald C. Fluctuations in syringe-pump infusions: association with blood pressure variations in infants. Am J Health Syst Pharm 1995;52:164653.
5. Dunster KR, Colditz PB. Flow continuity of infusion systems at low flow rates. Anaesth Intensive Care 1995;23:6059.
6. Rakza T, Richard A, Lelieur AC, Villaume I, Huyghe A, Nempont C, Storme L. [Factors altering low-flow drug delivery using syringe pumps: consequences on vasoactive drug infusion in preterm infant] [in French]. Arch Pediatr 2005;12:54854.
7. van der Eijk AC, van Rens RM, Dankelman J, Smit BJ. A literature review on flow-rate variability in neonatal IV therapy. Paediatr Anaesth 2013;23:921.
8. Rooke GA, Bowdle TA. Syringe pumps for infusion of vasoactive drugs: mechanical idiosyncrasies and recommended operating procedures. Anesth Analg 1994;78:1506.
9. Weiss M, Neff T, Gerber A, Fischer J. Impact of infusion line compliance on syringe pump performance. Paediatr Anaesth 2000;10:5959.
10. Neal D, Lin JA. The effect of syringe size on reliability and safety of low-flow infusions. Pediatr Crit Care Med 2009;10:5926.
11. Jonckers T, Berger I, Kuijten T, Meijer E, Andriessen P. The effect of in-line infusion filtering on in-line pressure monitoring in an experimental infusion system for newborns. Neonatal Netw 2014;33:1337.
12. Boehne M, Jack T, Köditz H, Seidemann K, Schmidt F, Abura M, Bertram H, Sasse M. In-line filtration minimizes organ dysfunction: new aspects from a prospective, randomized, controlled trial. BMC Pediatr 2013;13:21.
13. Jack T, Boehne M, Brent BE, Hoy L, Köditz H, Wessel A, Sasse M. In-line filtration reduces severe complications and length of stay on pediatric intensive care unit: a prospective, randomized, controlled trial. Intensive Care Med 2012;38:100816.
14. van Lingen RA, Baerts W, Marquering AC, Ruijs GJ. The use of in-line intravenous filters in sick newborn infants. Acta Paediatr 2004;93:65862.
15. Foster JP, Richards R, Showell MG, Jones LJ. Intravenous in-line filters for preventing morbidity and mortality in neonates. Cochrane Database Syst Rev 2015;8:CD005248.
16. Mirtallo J, Canada T, Johnson D, Kumpf V, Petersen C, Sacks G, Seres D, Guenter P. Task force for the revision of safe practices for parenteral nutrition. Safe practices for parenteral nutrition. J Parenter Enteral Nutr 2004;28:S3970Erratum in J Parenter Enteral Nutr 2006;30:177.
17. Bethune K, Allwood M, Grainger C, Wormleighton C; British Pharmaceutical Nutrition Group Working Party. Use of filters during the preparation and administration of parenteral nutrition: position paper and guidelines prepared by a British pharmaceutical nutrition group working party. Nutrition 2001;17:4038.
18. Lovich MA, Pezone MJ, Maslov MY, Murray MR, Wakim MG, Peterfreund RA. Infusion system carrier flow perturbations and dead-volume: large effects on drug delivery in vitro and hemodynamic responses in a swine model. Anesth Analg 2015;120:125563.
19. Lovich MA, Wakim MG, Wei A, Parker MJ, Maslov MY, Pezone MJ, Tsukada H, Peterfreund RA. Drug infusion system manifold dead-volume impacts the delivery response time to changes in infused medication doses in vitro and also in vivo in anesthetized swine. Anesth Analg 2013;117:13138.
20. Tsao AC, Lovich MA, Parker MJ, Zheng H, Peterfreund RA. Delivery interaction between co-infused medications: an in vitro modeling study of microinfusion. Paediatr Anaesth 2013;23:339.
21. Ma H, Lovich MA, Peterfreund RA. Quantitative analysis of continuous intravenous infusions in pediatric anesthesia: safety implications of dead volume, flow rates, and fluid delivery. Paediatr Anaesth 2011;21:7886.
22. Moss DR, Bartels K, Peterfreund GL, Lovich MA, Sims NM, Peterfreund RA. An in vitro analysis of central venous drug delivery by continuous infusion: the effect of manifold design and port selection. Anesth Analg 2009;109:15249.
23. Lovich MA, Kinnealley ME, Sims NM, Peterfreund RA. The delivery of drugs to patients by continuous intravenous infusion: modeling predicts potential dose fluctuations depending on flow rates and infusion system dead volume. Anesth Analg 2006;102:114753.
24. Parker MJ, Lovich MA, Tsao AC, Wei AE, Wakim MG, Maslov MY, Tsukada H, Peterfreund RA. Computer control of drug delivery by continuous intravenous infusion: bridging the gap between intended and actual drug delivery. Anesthesiology 2015;122:64758.
Copyright © 2016 International Anesthesia Research Society
25. Lovich MA, Doles J, Peterfreund RA. The impact of carrier flow rate and infusion set dead-volume on the dynamics of intravenous drug delivery. Anesth Analg 2005;100:104855.