Can a low prime volume arterial filter be used as an alternative for a venous bubble trap in minimal extracorporeal circulation? An in vitro investigation

Abstract

Background

During cardiac surgery the use of a minimal extracorporeal circulation (MiECC) system may reduce the adverse effects for the patient. This is probably caused by reduced inflammation and hemodilution. For the use of a MiECC circuit, a venous bubble trap (VBT) is warranted for safety reasons. The aim of this study was to assess if an arterial filter with a small prime volume has the same (or better) air removal capacities as a VBT in a MiECC circuit and subsequentially may be used as an alternative.

Methods

In an in vitro study, air removal properties were compared between the arterial filter and three VBT’s on the market, VBT160 (Getinge), VBT 8 (LivaNova and VARD (Medtronic). In a MiECC circuit, the filter devices were placed in a venous position and challenged with massive and micro air. Gaseous microemboli (GME) were measured with a bubble counter proximal and distal of the VBT device.

Results

More than 99.9 % of the air was removed after a bolus air challenge by all VBT’s. Both the VARD and the AF100 showed better GME removal properties (not significant for the AF100) compared to the other devices. All filters showed GME generation after a challenge with massive air. Compared to the other filters, only the VARD showed no passing of larger bubbles when a volume of 50 mL of air was present in the filter.

Conclusions

The AF100 seems to be a safe and low prime alternative for use in a MiECC system as a venous air trap. A word of caution, placement of the AF100 arterial filter in the venous line is off label use.

Keywords

cardiopulmonary bypass gaseous micro emboli venous bubble trap gaseous microemboli air extracorporeal circulation minimal extracorporeal circulation arterial filter

Introduction

During cardiac surgery the use of a minimal extracorporeal circulation (MiECC) circuit may reduce the adverse effects for the patient. A meta-analysis of Anastasiadis et al.¹ showed that the mortality rate of heart surgery was improved using MiECC, which is probably caused by reduced inflammatory response and less hemodilution.

Prime volume is the amount of volume needed to fill the extracorporeal circuit (ECC), typically 1.5-2 L, and is considered as a major contributor to hemodilution, because this volume is relatively large compared to the blood volume of the patient, which for an average patient is approximately 5 L. Ranucci et al. show in two studies that even a small reduction in prime volume may protect the patient from red blood cell transfusion² and may also affect goal directed perfusion strategies because hemoglobin levels are a major contributor to oxygen delivery to the patient.³ In the latter study they show that an 1% decrease of nadir hematocrit during cardiopulmonary bypass (CPB) leads to a 7% increase in the relative risk of acute kidney injury. The reduced prime volume of a MiECC system (approximately 1 L) to a conventional ECC, especially in combination with autologous priming may contribute more to decreased hemodilution.

The reduction in hemodilution in a MiECC circuit is achieved by, amongst other techniques, the removal of the venous reservoir resulting in a closed circuit. For the use of a closed CPB circuit, a venous bubble trap (VBT) is warranted for air removal and safety reasons.⁴ Air entering the MiECC system should be trapped by this filter and subsequently evacuated before reaching the arterial pump, the oxygenator, and, possibly, even the patient.

Various VBT’s are available on the market and their prime volumes vary between 128 and 212 mL (Table 1). Some VBT’s are designed similar to the arterial filter concept. We hypothesize that an arterial filter with a prime volume of 100 mL located in the venous line may be an interesting alternative as a VBT, because the small volume and the good air removal properties.⁵ It should be noted that the use of an arterial filter in the venous position of a MiECC system is considered as off label use.

Table 1.

Characteristics of venous bubble filters included in this study and the AF100 arterial filter. The AF100 seems a safe alternative compared to devices which are specifically developed as a venous air filter.

	Volume (mL)	Pore size (µm)
Inspire VBT 8	128	38
Affinity VARD^a	212	38
AF100	100	30
VBT 160	160	175

(VARD = venous air removal device).

^aThe VARD filter was used with the automatic air removal system.

The aim of this study was to assess if an arterial filter with a small prime volume has the same (or better) air removal capacities as a VBT in a MiECC circuit and subsequentially may be used as an alternative.

The air handling properties of the small prime volume arterial filter will be compared with the air handling properties of three other VBT’s.

Materials and methods

Design

A CPB test circuit was set-up based on a MiECC circuit and a pseudopatient (Figure 1). The MiECC circuit consisted sequentially of the tested air removal device in the venous line (Table 1), the VBT 160 (Bioline coated, Getinge, Rastatt, Germany), the VBT 8 (Phisio coated, LivaNova, Mirandola, Italy), the VARD (Venous Air Removal Device, Medtronic, Minneapolis, MN, USA) with the automatic air removal system (AAR1000 controller, Medtronic), or the Affinity AF100 (Medtronic). The circuit was then followed by a BPX80 Bio-pump (Medtronic) centrifugal pump and an Affinity Fusion oxygenator (Medtronic), with a gas blender setting of 2 L/min gas flow and with a fraction of delivered oxygen of 45%, resulting in a pO₂ of 37-38 mmHg, followed by the pseudopatient. All Medtronic products were Balance^® coated. The pseudopatient which functioned as a patient (volume) and as an air trap consisted sequentially of an AF100 (Medtronic) and an Affinity venous hardshell reservoir (Medtronic) which drained into the MiECC circuit. All tubing consisted of 3/8 inch and was Balance^® (Medtronic) coated.

Bovine blood was obtained from the slaughterhouse and no animal research approval was needed. Bovine blood (diluted with Ionolyte (Fresenius Kabi, Huis ter Heide, Netherlands) to a hematocrit of approximately 25 %) was pumped with 4.5 L/min and flow was measured with an external flow probe (SonoTT, EMTEC GmbH, Finning, Germany). Blood temperature was maintained at 32°C with a heater (Julabo GmbH, Seelbach, Germany). Venous line pressure was adjusted with a Hoffman screw clamp to approximately −25 mmHg and the arterial line pressure was 125 mmHg.

Three sample devices of each brand of VBT and AF100 were tested.

Air challenges

The following air challenges were divided in bolus, gaseous microemboli (GME) and breakthrough challenges. The various in vitro challenges were setup to mimic, under controlled conditions, the different challenges which may occur in the clinical setting. The bolus and GME challenges are described earlier.^6,7 A challenge was only started when the baseline showed a GME activity of <2 bubbles/s.

Bolus challenge

Using a 10 mL syringe on a three-way stopcock, 5 mL of air was quickly introduced (approximately 1 s) 40 cm proximal to the test VBT or AF100 (see Bolus challenge position in Figure 1). The collected air in the VBT or AF100 was removed within 5-10 s through the top purge port with a syringe. Air in the VARD filter was automatically removed with the Active Air Removal Controller (AAR1000, Medtronic). The controller was activated by the internal air sensors of the VARD.

The GME data were obtained from sensor 2 (see Figure 1 for position) and captured during a window of 60 s. This test was performed 5 times for each tested device.

Gaseous microemboli challenge

Air was introduced with a roller pump at 200 mL/min through ¼ inch tubing via a three-way stop cock in the venous line circa 40 cm proximal to the hardshell reservoir of the pseudopatient. This resulted in a continuous GME activity leaving the hardshell reservoir and challenging the tested filters. After reaching a constant GME load, GME was measured for 2 min. GME entering the device were measured with the bubble sensor 1 and the GME behind the device were measured with sensor 2 (see Figure 1 for the positions). From the measurement of both sensors the reduction rate could be calculated.

This test was conducted one time for each tested device. Meaning that the test was repeated three times, because three devices per brand were used for the study.

Breakthrough threshold

Using a 50 mL syringe placed on a three-way stop cock 40 cm proximal to the test device, a total of 50 mL of air was introduced in increments of 10 mL, 15 and 25 mL. The air trapped in the VBT or AF100 device was not evacuated. GME were measured continuously distal to the filter with sensor 2 (see Figure 1 for the position). This test was performed one time for each tested device. Meaning that the test was repeated three times, because three devices per brand were used for the study.

Gaseous micro emboli

GME were detected using the bubble counter BCC 300 (GAMPT mbH, Zappendorf, Germany). The measurements were conducted with two non-invasive sensor probes. The probes were clamped onto the venous line 5 cm distal to the venous hardshell reservoir and on the in-flow tubing of the oxygenator device (Figure 1). The BCC 300 accurately measures the number and the size of GME, with a diameter ranging from 20 to 2000 μm, using Doppler ultrasound (2 MHz). The device specifies the bubbles with a diameter greater than 2000 μm as ‘over range’.

Statistics

Quantitative variables are presented as mean ± standard deviation (SD) or as median with Interquartile ranges (IQR), when appropriate. We repeated five bolus tests for every tested device. Of these five tests the average was calculated and this value was considered as n = 1. The mean was calculated from the averages obtained from the same devices. Before analysis, data were tested for distribution according to Kolmogorov–Smirnov goodness of fit test. ANOVA or Kruskal–Wallis tests were used to assess statistical significant differences for the various groups. A p-value <.05 was considered significant. The Student’s t-test or Mann-Whitney tests were used, when appropriate, for comparisons focusing on two groups. Significance was reached according Bonferroni corrected p-values.

Statistical analysis was performed using SPSS 22.0 (SPSS, Chicago, IL, USA).

Results

Bolus challenge

The results of the bolus challenge of the various VBT and the AF100 devices are shown in Figure 2. An ANOVA test comparing the four groups showed a significant difference between the groups (p = .044). Comparing two groups showed a significant difference between the VARD filter (0.18 ± 0.07 µL) and both the VBT 8 and the VBT 160 (Volume: 0.88 ± 0.15 µL and 1.51 ± 0.44 µL; p = .002 and p = .007, respectively). No significant differences were found for pairwise comparison between the other devices, including the AF100 (1.11 ± 0.80 µL.

Gaseous micro emboli challenge

The results of the GME challenge are shown in Figure 3. Between the four groups a significant difference in GME reduction was shown (ANOVA; p = .019). Testing between two groups showed significant differences between the VBT 8 and the VBT 160 (respectively, Volume: 14 ± 26% and 9 ± 25%; p = .008 and .006) compared to the VARD filter (88 ± 5%) only. No significant differences were detected between the AF100 (61 ± 40%) and the other devices.

Breakthrough threshold

The results of the breakthrough challenge are shown in Table 2. A breakthrough threshold was defined as spilling of air bubbles larger than 2000 µm (measured by the bubble counter as ‘over range’). The VBT 8 filter showed in two of the three filters a breakthrough of larger air bubbles with a total of 25 mL air in the filter, while the other filters showed no breakthrough at this volume. Figure 4 shows an AF100 filter after the introduction of a total of 25 mL of air. All filters showed a breakthrough threshold for all three tested devices at a total volume of 50 mL with the exception of the VARD filter, which showed no breakthrough phenomenon in all tests.

Table 2.

Observation of breakthrough threshold of every individually tested VBT or AF100 (n = 3) with cumulative air volume.

	Cumulative air volume (mL)
	10	25	50
Affinity VARD	GGG	GGG	GGG
VBT 8	GGG	BBG	BBB
VBT 160	GOG	GGG	GBB
AF100	GGG	GGG	BBB

O = no GME detected distal to device; G = Continuous GME detection distal to device; B = Breakthrough (air larger than 2000 µm was observed).

(VBT = venous bubble trap; VARD = venous air removal device; GME = gaseous microemboli).

Discussion

In this in vitro study an AF100 arterial filter was positioned in the venous line of a MiECC circuit and air removal properties were compared with three filter devices which were specifically developed as a venous air filter. The filters were challenged with various air challenges; macro air (bolus), micro air (GME) and massive air. We showed that the air removal properties of the AF100 arterial filter were good and comparable to the three VBT’s.

GME originate from the CPB/MiECC system and are caused by actions required during CPB, such as blood sampling, transfusing volume or injecting drugs.^8,9 Besides, introduction of air into the venous line of the CPB circuit due to residual air in the venous cannula after connecting the tubes or by the excessive negative pressure will always lead to a detectable amount of GME in the arterial line.^10–12

These GME are thought to contribute to the adverse outcomes of cardiac surgery.¹³ The blood-air surface of GME can activate the inflammatory response, may damage the endothelium and obstruct the blood flow in the capillary vessels, causing transient ischaemia of end-organs.¹⁴

A bolus air challenge was extremely well removed by all VBT’s and also by the arterial filter, AF100. A maximum of 1.5 µL was observed behind the tested filters after a bolus challenge of 5 mL, meaning that more than 99.97 % of the introduced air was removed. The VARD filter showed the best air removal properties of this bolus challenge. The automatic air removal probably added to this effect, since the air was quickly removed from the top of the filter therefore avoiding spilling of GME. In the other filters we choose to remove the collected air after approximately 10 s, which is equal to a clinical setting wherein the perfusionist manually removes air after observing visual air in the top of the filter. Large air volume in the filter will always lead to spilling of GME, as we showed in the breakthrough threshold test.

Interestingly, the filters without automated air removal showed similar results despite large differences in design. Air-removal by contemporary bubble filters is accomplished by a complicated combination of bubble buoyancy (influenced by velocity and centrifugal forces) and mechanical screen filtration.¹⁵ It seems that reduction of larger air, as we did in the bolus challenge, was not dependent on pore size and volume of the filter. It is likely that the pressure necessary for the bubble to pass across the screen barrier, the bubble point pressure in our controlled study design, was not exceeded.¹⁶

Both the VARD and the AF100 showed better GME removal properties (although not significant for the AF100) compared to the other devices. This is likely caused by the design of both filters. Both the VARD (without automated air removal device) and the AF100 were developed as an arterial filter with a small screen pore size (38 and 30 µm, respectively) and a larger surface area is created by folding of the filter material. This is also true for the VBT 8, however, this filter did not function better.

Detailed GME measurements in a previous study⁴ showed that in MiECC system in a clinical setting a VBT 160, with a screen pore size of 160 µm, had a cut-off of approximately 400 µm, meaning that almost all GME larger than 400 µm were removed. We showed in another study that arterial filters with screen pore sizes between 25 and 40 µm have lower cut off values varying between 100 and 160 µm.¹⁷ We may expect that the positioning of an arterial filter in the venous side of a MiECC system contributes more strongly to air removal which is supported by the data from our study. A word of caution, when massive air fills a filter with small screen pore size this filter is probably very difficult to deair.

We showed that the VBT 8 was the only filter spilling large bubbles at 25 mL of cumulative air. Both the VBT 160 and the AF100 started spilling larger air bubbles when 50 mL of air was present in the filter. Only the VARD showed no large over range bubbles distal to the filter. Possibly, as we also showed with the bolus challenge, the properties of the screen filter (i.e. surface tension, surface size and pore size), are the major contributors to avoiding air breakthrough.

Overall, the air removal properties of the AF100 arterial filter when placed in the venous line, are very good and comparable with devices which are specifically developed as a venous air filter. The AF100 seems to be a good GME scavenger, possibly reducing adverse effects of CPB. Additionally, the properties of removing or blocking massive air for the AF100 are similar to those of the venous filters, thus protecting the patient and adding to safety. According to ‘Guidance for Extracorporeal Blood Circuit Defoamer 510(k)’ of the FDA, in which testing for market approval of arterial and venous air traps are described, new devices should be equivalent in air removal compared to predicate devices. We believe that our study proves that the AF100 has equivalent air removal properties compared to predicate VBT’s. Accordingly, the AF100 seems to be a safe alternative, and this is achieved with the lowest prime volume. Since hemodilution is considered as a major contributor to the detrimental effects of CPB, the low prime volume of this filter can be considered as beneficial. Although the prime volume difference is low, especially in a CPB set-up for MiECC the AF100 may be a good option.

A limitation of this study may be that the air introduction was generated in vitro with a strict separation between a GME and a bolus challenge. The air challenge in a clinical setting may be a combination of both, which may influence the air removal properties of the devices.

We decided to measure behind the centrifugal pump, with the idea that this pump fractionates the bubbles and concordantly improving the measurement with the bubble counter. However, possibly the centrifugal pump also has air removal properties influencing the measured GME volume.

We compared the devices with volume measurements and did not specify distribution of the diameters. We know that the distribution was in the lower diameter range. GME volume measured after a bolus challenge was not higher than 1.5 µL, while a diameter of 2 mm (over range) will give a bubble of 4 µL. No over range bubbles were detected after the bolus and GME challenges.

Importantly, placement of the AF100 arterial filter in the venous line is off label use. According to the instructions for use by Medtronic the product should be positioned in the arterial line.

Conclusion

The air removal properties of the AF100 filter in the venous position are very good and comparable with devices which are specifically developed as a venous air filter. Both massive air and GME were scavenged equally well or better by the AF100 compared to the other filters.

In conclusion, the AF100 seems to be a safe and low prime alternative for use in a MiECC system as a venous air trap. A word of caution, placement of the AF100 arterial filter in the venous line is off label use.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by an unrestricted grant of Medtronic.

ORCID iD

Marco Stehouwer

References

Anastasiadis

Antonitsis

Haidich

, et al. Use of minimal extracorporeal circulation improves outcome after heart surgery; a systematic review and meta-analysis of randomized controlled trials. Int J Cardiol 2013; 164(2): 158–169.

Ranucci

Pistuddi

Carboni

, et al. Effects of priming volume reduction on allogeneic red blood cell transfusions and renal outcome after heart surgery. Perfusion 2015; 30(2): 120–126.

Ranucci

Aloisio

Carboni

, et al. Acute kidney injury and hemodilution during cardiopulmonary bypass: a changing scenario. Ann Thorac Surg 2015; 100(1): 95–100.

Roosenhoff

Stehouwer

De Vroege

, et al. Air removal efficiency of a venous bubble trap in a minimal extracorporeal circuit during coronary artery bypass grafting. Artif Organs 2010; 34(12): 1092–1098.

Stehouwer

Boers

de Vroege

, et al. Clinical evaluation of the air removal characteristics of an oxygenator with integrated arterial filter in a minimized extracorporeal circuit. Int J Artif Organs 2011; 34(4): 374–382.

Stehouwer

de Vroege

. Air removal capacity of two different minimal invasive ECC systems: an in vitro comparison. Perfusion 2019; 34(7): 561–567.

Stehouwer

Kelder

van Oeveren

, et al. In vitro air removal characteristics of two neonatal cardiopulmonary bypass systems: filtration may lead to fractionation of bubbles. Int J Artif Organs 2014; 37(9): 688–696.

Taylor

Borger

Weisel

, et al. Cerebral microemboli during cardiopulmonary bypass: increased emboli during perfusionist interventions. Ann Thorac Surg 1999; 68(1): 89–93.

Borger

Feindel

. Cerebral emboli during cardiopulmonary bypass: effect of perfusionist interventions and aortic cannulas. J Extra Corpor Technol 2002; 34(1): 29–33.

10.

Aboud

Liebing

Borgermann

, et al. Excessive negative venous line pressures and increased arterial air bubble counts during miniaturized cardiopulmonary bypass: an experimental study comparing miniaturized with conventional perfusion systems. Eur J Cardio Thorac Surg 2014; 45(1): 69–74.

11.

Simons

Ganushchak

Teerenstra

, et al. Hypovolemia in extracorporeal life support can lead to arterial gaseous microemboli. Artif Organs 2013; 37(3): 276–282.

12.

Rodriguez

Rubens

Belway

, et al. Residual air in the venous cannula increases cerebral embolization at the onset of cardiopulmonary bypass. Eur J Cardio Thorac Surg 2006; 29(2): 175–180.

13.

Barak

Katz

. Microbubbles: pathophysiology and clinical implications. Chest 2005; 128(4): 2918–2932.

14.

Muth

Shank

. Gas embolism. N Engl J Med 2000; 342(7): 476–482.

15.

Fiore

Morbiducci

Ponzini

, et al. Bubble tracking through computational fluid dynamics in arterial line filters for cardiopulmonary bypass. Am Soc Artif Intern Organs J 2009; 55(5): 438–444.

16.

De Somer

. Evidence-based used, yet still controversial: the arterial filter. J Extra Corpor Technol 2012; 44(1): 27–30.

17.

Stehouwer

Legg

De Vroege

, et al. Clinical evaluation of the air-handling properties of contemporary oxygenators with integrated arterial filter. Perfusion 2017; 32(2): 118–125.