Keywords

1 Introduction

The complications associated with indwelling ureteral stents, namely bacterial adhesion and biofilm formation, have been the main driving force for the development of new materials or coatings with antimicrobial and anti-adhesive properties. The first approach for testing and optimizing new biomedical surfaces usually consists of evaluating their in vitro efficacy under controlled experimental conditions that reflect the human physiological environment [1]. Consequently, several parameters, including the pathogenic species and their concentration, culture medium, temperature, and hydrodynamic conditions, must be considered when setting an in vitro experiment, hence increasing its predictive value and avoiding, during initial screening, expensive in vivo assays and animal sacrifice [1] without prior evidence of surface effectiveness. Among these parameters, hydrodynamic conditions have a prominent role in the experimental setup as assays performed in static conditions do not mimic the fluid flow that occurs at specific locations of the human body (e.g. urinary tract). Furthermore, it is well known that hydrodynamic conditions affect not only bacterial adhesion to biomedical surfaces [2], but also biofilm growth and architecture [3, 4]. In fact, flow determines the transport rate of planktonic cells to the surface and their subsequent interaction [5], as well as the transport of oxygen and nutrients to the biofilm [6]. Besides, flow influences both bacterial attachment and detachment rates [7].

The effectiveness of biomedical surfaces may also be highly affected by the hydrodynamic conditions [1]. Surfaces releasing antimicrobial substances when exposed to flow may exhibit shorter lifetimes than at static conditions [1]. Likewise, depending on the fluid flow surrounding the surface, contact-killing surfaces that are adhesive for bacterial cells may be covered by bacterial debris, which decreases their antimicrobial activity [1]. Lastly, non-adhesive coatings, such as polymer brush coatings, are generally sensitive to external stimuli, exhibiting higher antifouling performance at quasi-static conditions and more effective fouling release behavior under dynamic conditions [8].

Considering the importance of hydrodynamic conditions and their effects on bacterial adhesion and biofilm formation, a diversity of in vitro flow systems, including the Robbins device (RD) and modifications, the drip flow biofilm reactor, rotary biofilm reactors and flow chambers (FCs), have been developed and optimized to evaluate surfaces effectiveness under physiological conditions [9]. Certain flow systems enable real-time visualization of bacteria adhesion/biofilm development under controlled conditions (e.g. shear stress or shear rate, temperature), allow simultaneous testing of different materials, and can be used as high-throughput platforms [9], while others have some limitations in operating at highly controlled hydrodynamic conditions [1]. Hence, each platform presents advantages and disadvantages that must be considered before use.

In this chapter, the most commonly used platforms for the in vitro assessment of bacterial adhesion and biofilm formation under flow conditions—the modified Robbins device, flow chambers, and microfluidic devices—are introduced, and their main advantages and disadvantages discussed. These three testing platforms have been particularly used to evaluate the anti-adhesive and antibiofilm performance of novel surface materials for urinary tract devices (UTDs), including catheters and stents, due to their ability to control the hydrodynamics (shear stress and flow rate) and recreate in vivo flow conditions.

2 Robbins Device and Modifications

The Robbins device was initially developed by Jim Robbins and Bill McCoy to study biofilm formation in industrial water systems [10]. The RD consists of a pipe with several holes where coupons are mounted on the end of the screws and become in contact with the fluid. Thus, the RD generates submerged biofilms growing in aqueous systems that can be used for the investigation of multispecies communities [10].

Several modifications were later introduced to this design, including the use of a square-channel pipe where coupons are aligned with the inner surface without disturbing flow characteristics [11]. Other designs include a half-pipe geometry that more closely resembles the circular section of a tube [4]. With the modified Robbins devices (MRDs), the flow can be momentarily stopped to allow direct access to the coupons so that time-course experiments are also possible [3].

MRDs have been operated in conditions that mimic the flow in urinary catheters [12, 13] and stents [13, 14]. Tunner et al. [14] were among the first authors to use a continuous flow model based on an MRD to assess encrustation on silicone and polyurethane, the most widely used ureteral stent biomaterials. They revealed that the type and degree of encrustation produced were similar to those found in vivo, recommending this flow system for comparative evaluation of surface candidates for medical devices used in the urinary tract [14]. More recently, in our research group, a MRD (referred to as flow cell system) simulating the hydrodynamic conditions found in urinary catheters (shear rate of 15/s) [15] was used to characterize the microbial physiology of Escherichia coli and Delftia tsuruhatensis individually and in a consortium, in terms of growth kinetics and substrate uptake, when exposed to artificial urine medium (AUM) flow and silicone material [12]. Additionally, we used a custom-made semi-circular flow cell identical to that shown in Fig. 1 to assess the efficacy of different nanocomposite coatings in preventing urinary tract infections (UTIs) [13]. The hydrodynamics of this flow cell was fully characterized by computational fluid dynamics (CFD) [16], and it has been shown that the shear stress field is approximately the same in the curved and flat walls so that coupons can be placed on the flat wall for convenience and still be subjected to the same shear forces acting on the curved wall [17]. Moreover, this flow cell was constructed to have enough inlet length to allow for full flow development and a large surface area on which the hydrodynamic conditions remain constant for a wide range of flow velocities [16]. These dynamic systems are particularly useful for screening purposes as they enable the simultaneous testing of several surfaces [13, 14]. Another advantage of MRDs is that coupons can be removed independently, for instance, at different experimental times [12].

Fig. 1
figure 1

(a) Schematic representation and (b) photograph of a MRD. The system is mainly composed by a recirculating tank, one vertical semi-circular flow cell (about a meter high) with removable coupons, and peristaltic and centrifugal pumps

Full size image

3 Flow Chamber

Despite the many advantages of the MRDs, they are usually not suited for direct analysis of biofilm development [18], and they are not adequate to monitor cell adhesion to a surface. Nowadays, there are several models of flow chambers that can be mounted on a microscope stage and used with video capture systems, enabling real-time observation of microbial adhesion, particularly when used with transparent surfaces [18]. Different custom-made FCs have been used to evaluate the anti-adhesive and antibiofilm properties of novel surfaces for UTDs, namely catheters and stents, in flow conditions that simulate those typically found in these medical devices [2, 15, 19, 20]. Table 1 summarizes several studies found in the literature where flow chamber assays were performed under fully characterized hydrodynamic conditions similar to those of urinary catheters and stents. Most of these studies aimed to monitor the initial adhesion of bacteria associated with UTIs (E. coli, Enterococcus faecalis, Staphylococcus aureus and Pseudomonas aeruginosa) to polymeric surfaces as polydimethylsiloxane (PDMS) [2, 9] and PDMS modified with antimicrobial substances (peptides and carbon nanotubes) [21,22,23] for 30 min to 4 h. In some instances, these systems were also used to investigate bacterial biofilm growth and survival for 24 h on novel surface coatings for UTDs [19, 24, 25].

Table 1 Flow chamber studies to evaluate the initial adhesion and biofilm formation under hydrodynamic conditions identical to those found in UTDs
Full size table

A custom-made FC system (Fig. 2) was designed by our group to analyse cell adhesion [22, 26] and biofilm formation [19, 24]. This system includes a parallel-plate flow chamber (PPFC) coupled to a jacketed tank and connected to centrifugal pumps and a valve by a silicone tubing system. The valve allows the bacterial suspension to circulate through the system at a controlled flow rate, and the recirculating water bath is connected to the tank jacket to enable temperature control. To illustrate the type of data that can be obtained with this platform, biofilm formation experiments with E. coli were carried out for 24 h using PDMS as the test surface [27] and AUM recirculated through the FC system at 4 mL/s to mimic the urine flow behavior in ureteral stents (shear rate of 15/s). After 24 h, the system was stopped, and the biofilm formed on the PDMS surface was stained with a fluorescent dye and analysed by confocal laser scanning microscopy (CLSM) (Fig. 3 and Table 2).

Fig. 2
figure 2

(a) Schematic representation and (b) photograph of the FC system. The PPFC is coupled to a glass tank connected to four centrifugal pumps and a tubing system to conduct adhesion or biofilm formation assays

Full size image
Fig. 3
figure 3

3-D projection of biofilms formed on PDMS at a flow rate of 4 mL/s mimicking ureteral stents in the described PPFC system. Shown is an E. coli biofilm stained with SYTO 61 (633 nm laser line, LEICA HCX PL APO 10 ×/0.40 CS). This representative image was obtained using the “Easy 3D” tool of IMARIS 8.4.1 software (Bitplane, Switzerland) from a confocal z stack, and presents an aerial view of the biofilm structure with the shadow projection on the right

Full size image
Table 2 Quantified data for E. coli biofilms grown on PDMS surfaces in the PPFC system. These parameters were obtained from confocal image series using the COMSTAT2 tool associated with the ImageJ software. The means (± standard deviations) for three independent experiments are presented
Full size table

CLSM is an optical imaging technique used to obtain high-resolution images of biofilms at various depths in their naturally hydrated form and to generate three-dimensional (3-D) reconstructions of the samples [28]. It is particularly well suited for monitoring 3-D structure formation in flow chamber-grown biofilms due to its non-invasive and non-destructive character [29, 30]. Early research investigating the use of CLSM in biofilm studies was more descriptive, using qualitative metrics to evaluate biofilm architecture [31]. The development of imaging software packages, specifically for biofilm samples, has enhanced the quantitative output from CLSM images of biofilms [32]. Among these, the COMSTAT ImageJ plugin [32] used in the present work (Table 2) or the PHLIP Matlab toolbox [33, 34] represent a set of reference tools that are efficient and reliable to characterize biofilms in terms of biomass, thickness distribution, surface coverage, roughness coefficient, or porosity.

4 Microfluidic Devices

Microfluidic platforms have demonstrated high potential and versatility for the study of bacterial adhesion and biofilm formation under different growth conditions. These platforms allow the testing of different channel architectures and types of materials or surfaces at highly controlled flow conditions through a rapid and precise analysis [5]. For these reasons, microfluidic platforms have been used to explore the combined effect of several factors on the development of clinically relevant biofilms [35,36,37]. Table 3 lists several studies using microfluidic devices for the evaluation of bacterial adhesion and biofilm formation under flow conditions that represent relevant hydrodynamic regions of ureteral stents.

Table 3 Microfluidic platforms used for the study of bacterial adhesion and biofilm formation under hydrodynamic conditions identical to those found in ureteral stents
Full size table

Although microfluidic devices can be constructed by different methodologies and from a diversity of materials, PDMS has been the material of choice for the construction of these devices, with most of the PDMS-based microfluidic devices being designed for a specific purpose. Several studies have investigated the initial bacterial adhesion on different materials using microfluidic platforms [5, 7, 38,39,40,41]. In general, the bacterial residence time and surface coverage increased linearly up to 3.5 Pa [7] and 20/s [40], respectively, and the adhesion rates were higher in locations with a sudden increase in shear forces [39]. For the particular case of ureteral stents, De Garcia et al. [5] demonstrated that unobstructed devices (wall shear stress ≤ 0.0875 Pa) showed no short-term bacterial adhesion, while in obstructed devices, the cavity region and nearby proximal side-hole (wall shear stress of 0.131–0.175 Pa) exhibited higher levels of bacterial attachment compared to other regions of the model. Although channel architecture and geometry affect bacterial adhesion [41], these findings indicate that flow influences both attachment and detachment rates [7].

PDMS-based microfluidic devices have also been applied to explore how bacterial colonization, competition, and dispersal occur at flow conditions. Indeed, flow can confer growth advantages to pathogens by allowing the bacteria upstream movement [42]. Similarly, the study of biofilm development is also possible using these microfluidic platforms [35, 43,44,45,46,47,48]. Several authors revealed that flow alone was able to induce the formation of polysaccharide intracellular adhesins [46] and was the major modulator of the biofilm structures [45]. Additionally, Lee et al. [35] demonstrated that the morphology of Staphylococcus epidermidis biofilm formation was influenced by local hydrodynamic conditions. While higher wall shear stress limited vertical biofilm growth, resulting in a monolayer structure, cells growing in stagnant areas were able to proliferate rapidly, resulting in the formation of a large multilayer structure [35]. Likewise, biofilm thickness was also affected by flow after 48 h, increasing significantly at 0.010 Pa (36 ± 9 μm) and slightly at 0.0035 Pa (20 ± 4 μm). Contrarily, no increase was detected for higher shear stresses [44]. Accordingly, Kim et al. [48] revealed that quorum sensing-mediated communication during biofilm formation was generally repressed by flow, impairing biofilm growth. The comprehensive analysis of gene expression during S. aureus biofilm formation was successfully conducted by Moormeier et al. [49, 50] using a different microfluidic device, the BioFlux system (Fluxion Systems, South San Francisco, CA), and compared with static conditions. The BioFlux system was presented as the most prominent commercial microfluidic platform that overcomes the limitations of static well plates and conventional laminar flow chambers. In this system, biofilm formation can be followed by light microscopy in microfluidic wells, allowing rapid screening of the effects of several compounds on the viability of biofilms under hydrodynamic conditions [51]. One of the early studies performed on this platform evaluated the effect of several antimicrobials on 8 h-developed P. aeruginosa biofilms under controlled hydrodynamic conditions at 37 °C. Results suggested that biofilm viability measured with the plate reader agreed with those determined using plate counts and with the results of fluorescence microscope image analysis. Since then, the BioFlux system has been considered a high-throughput methodology for the study of biofilm development under defined hydrodynamic conditions [36, 49, 50, 52,53,54].

Although only 1 of 21 analysed studies had the specific objective of evaluating bacterial adhesion in urinary stents, all provided a comprehensive analysis of adhesion and biofilm formation at flow conditions representative of relevant hydrodynamic regions of ureteral stents [55] and should be considered when testing a new surface or coating for these medical settings.

5 Operating Conditions

As previously shown, MRDs, flow chambers and microfluidic devices have been used to study bacterial adhesion and biofilm formation under hydrodynamic conditions that simulate the UTDs. Because the flow rate by itself provides little information about shear without taking into account the geometry of the in vitro flow system, it is crucial to mimic the flow conditions in a catheter or stent by using either the wall shear stress or the shear rate [1]. The wall shear rate (σ, with unit/s) is a measure of change of the fluid velocity near the wall of the tube in the radial direction toward the center of the tube. In laminar conditions, the shear rate is related to the force which the fluid flow exerts on the wall, expressed as shear stress (τ, with unit Pa), through τ = μ × σ, where μ is the dynamic viscosity of the fluid (10−3 Pa s for water). In the flow systems under study, the flow rate should be adjusted to approach an average shear rate of around 15/s as an estimate of the intraluminal urine flow, based on predictable daily urine production and internal catheter diameter [15]. Nevertheless, urinary output values are highly variable and may reach more than 10 times the mean value [56], yielding a proportional increase in wall shear rate. Some authors performed FC tests at a shear rate of 33/s, which is higher than mean values but still within the range of shear rates found in urinary catheters [25].

Regarding the flow chamber system described in this work (Fig. 2), the numerical simulations indicated that the shear rate of 15/s reported for urinary flow in catheters can be attained at a flow rate of 2 mL/s [2, 9]. On the other hand, the average shear stress in problematic zones of ureteral stents that are prone to encrustation (0.024 Pa) [55] can be obtained by operating the PPFC system at a flow rate of 4 mL/s [21]. In the case of MRDs used by our research group, the recirculation flow rates can range from 5 [12] to 53 mL/s [13] to mimic the shear forces on urinary catheters, depending on the geometry of the flow cell.

PDMS-based microfluidic devices are usually designed for a particular application, having their own architecture and geometry with specific operating conditions. In the case of the commercially available BioFlux system, numerical simulations revealed that the average shear stress value of 0.02 Pa reported for ureteral stents [55] can be reached at a flow rate of 66 μL/h [52].

6 Strengths and Limitations of Flow Platforms

Among the advantages of flow systems are the ability to compare, for instance, the effect that different substrates, media and hydrodynamic conditions exert on a biofilm at different developmental stages. These dynamic models may also provide an evaluation of the effect that transiently occurring molecules, such as antibiotics or adherence inhibitors, have on biofilms. However, the technical disadvantages of flow reactors include increased experimental complexity as well as possible formation/trapping of air bubbles in the setup tubing (particularly severe in microfluidic systems), as this can affect flow and biofilm architecture [57].

Choosing the experimental platform for flow experiments determines what kind of data can be extracted, and care must be taken to ensure that the selected reactor fulfills the objectives of the experiments. The three platforms covered in this chapter (modified Robbins device, flow chamber and microfluidics-based device) have benefits and limitations, which are summarized in Table 4.

Table 4 Advantages and disadvantages of dynamic biofilm cultivation devices
Full size table

7 Conclusions

To evaluate the anti-adhesive and antimicrobial performance of novel biomedical materials, a number of flow devices have been designed to recreate in vivo flow conditions. Shear stress and flow rate can be accurately controlled and varied in these in vitro flow systems, which requires prior knowledge of the flow dynamics inside the platform. After limiting their operational range, modified Robbins devices, flow chambers and microfluidic devices are suggested as experimental setups to mimic the flow behavior in urinary catheters and stents.