INTRODUCTION

Sometimes, different diseases or physical or chemical aggressions can lead to a loss or alteration of the cells of a tissue or organ. The normal restitution of this tissue is the main purpose of regenerative medicine ^[1]. Tissue engineering is a new area of regenerative medicine whose objective is the construction of in vitro tissues for a therapeutic use, that allows to restoring, replacing, or increasing the functional activities of the organic tissues themselves ^[2]. Basically, tissue engineering consists of growing cells in an enriched three-dimensional matrix, where these cells can grow and later be transplanted into a recipient organ ^[3]. Three components can be distinguished: cells, biomaterials (the extracellular matrix or supporting scaffolds) and growth factors. This paper will focus specifically on contributing information and methodologies in the construction of scaffolds for cartilaginous tissue. Among its possible applications is the aesthetic treatment of microtia, a congenital malformation of the external ear that ranges from mild structural anomalies to the complete absence of the ear, with a prevalence rate that varies significantly between countries ranging from 0.8 to 17.4 per 10,000 births, with a higher prevalence reported in Ecuadorians, Chileans and Finns, with prevalence of 17.4; 8.8 and 4.3 respectively for every 10,000 births ^[4]. Current cosmetic treatment options for children include reconstruction with prosthesis, combination of prosthesis and own tissue, and autologous rib cartilage transplant.

Scaffolds, which can be manufactured by 3D printing, are highly porous three-dimensional substrates that house cells, providing a surface to which they can adhere, thrive and multiply ^[5]; consequently, they generate an extracellular matrix (ECM) of structural and functional proteins and saccharides that form living tissue ^[6]. This type of structure has been used in different organs such as bone tissues or soft tissues ^[7]. During the last decades, much research has been carried out regarding structural capacity ^[8], biocompatibility, cell growth speed ^[9], shape and manufacturing process ^[10], among others. These investigations have generated extensive and interesting information that has made it possible to make considerable improvements in this discipline. More recently, diverse medical sub-specialties have started considering novel 3D bioprinting approaches, based on the concept of combining living cells and biomaterials, controlling cell proliferation, attachment, and migration within 3D printed scaffolds ^[11]. Currently, it is possible to design and manufacture cartilage in the shape of an ear, in vitro, which is subsequently implanted under the patient's skin. However, we must consider that these are new procedures and there is still a need to improve the methodologies and materials used. For this reason, it is important to define and evaluate manufacturing parameters to guarantee the manufacture quality of scaffolds. Different parameters such as material, geometry and pore size are evaluated, indicating how the manufacturing process can alter the expected result.

Two biocompatible and biodegradable materials, polycaprolactone (PCL) and polylactic acid (PLA), were used to manufacture the scaffolds. While there are other biocompatible materials, these two materials were used primarily because they are economic, although both present different difficulties in their manufacture. These materials will generate printing parameters that can be extrapolated to other biocompatible materials with similar characteristics.

MATERIALS AND METHODS

The steps needed to allow the characterization of the necessary parameters for 3D scaffolding printing will be developed in the following section.

Design and computational modeling of scaffolding

For the design of the PLA and PCL scaffolds, it is essential to define the following geometric parameters: l (scaffold length), p (pore size), h (scaffold height) and e (filament thickness). These parameters can be seen in Figure 1.

Figure 1 Top, three-dimensional view and dimensions (parameters) of the square scaffold. Source: self-made. 

Once these parameters were defined, it was necessary to define the type of pore geometry. In this work, the geometries shown in Tables 1 and 2 were evaluated. Table 1 considers the geometries manufactured with PLA and Table 2 considers the geometries manufactured with PLC.

Table 1 PLA scaffolding dimensions. 

Square	Triangular	Ellipsoidal
p1= 600 µm - b3: 300 µm p1= 450 µm - b2: 225 µm p1= 300 µm		ᶩ = 20 mm e = 200 µm h = 1,2 mm

Table 2 PCL scaffolding dimensions. 

Square
p1= 200 µm p1= 400 µm p1= 400 µm	ᶩ = 20 mm e = 400 µm h = 1,2 mm

The pore size is represented by p and b parameter. For square and triangular geometries three scaffolds were defined, for ellipsoidal geometry just two scaffolds were defined. Figure 2 shows the structure of triangular and ellipsoidal scaffolding.

Figure 2 Structure of triangular and ellipsoidal scaffolding. Source: self-made. 

RESULTS AND DISCUSION

During the scaffolding generation, the environmental conditions for printing process was analyzed. During the printing, it is necessary to consider the following aspects:

The printers were encapsulated in transparent acrylic boxes to prevent the entry of air currents that may impair the printing process and, at the same time, allow checking the status of each print without opening the box. They also decrease the possibility of entry of pathogens into the printing area.

A moisture absorbing filter were placed inside the box, as humidity greatly affects the properties of the materials which can, in turn, have an impact on the print quality.

The printers were manually calibrated to ensure proper adhesion of the material to the base and ensure that each print is level.

Glass was chosen as the printing surface to facilitate cleaning and hygiene.

For the printing process, special care was taken in the hygiene of all the implements and the surrounding area of the printing. For this, 70% (w/v) alcohol were used to disinfect the workspace and tools. This concentration is the classic and agreed recommendation of the reliable methods for decontaminating surfaces ^[20]. Additionally, it was contemplated the use of masks and clinical gloves.

This variable had a high impact on the scaffold's print quality. For the PLA case, the extruder temperature must be lowered to counteract the ambient temperature. If the ambient temperature exceeds 20 °C, the extruder temperature should be lowered beyond the limits recommended by the manufacturer. On the other hand, the PCL printing were more susceptible to variations on ambient temperature; when the temperature was below 10 °C, the printing was released by retraction because the material cools too quickly, and, if the temperature exceeded 20 °C, the material was kept in a viscous state for a long time, decreasing the pore size.

With consideration of those parameters, the following results are highlighted.

Manufacturing remarks for PLA

It is evidenced that it is not possible to manufacture designs with pores of 200 µm, because, to ensure adhesion to the glass base, the first layer of the print is wider, producing an effect that is commonly called "elephant foot". This causes the pore walls to join, which causes the pore to be "enclosed" at the bottom and not exposed on both sides of the scaffold (Figure 3).

Figure 3 Side view of square geometry scaffolding with "elephant foot" effect (not to scale). Source: self-made. 

It is possible to reduce this phenomenon by relying on Cura software ^[17]; however, this greatly hinders the adhesion of the material, which does not make its manufacture possible.

In the case of ellipsoidal scaffolding, it is evidenced that the way to print with precision, is to form each line with two thinner lines by using the option Special Modes-Surface: mode surface option ^[18] (Figure 4(A)) The normal option (used in the other geometries) doesn’t a satisfactory result (Figure 4(B)).

Figure 4 Ellipsoidal scaffolding. The images was taken from optical microscope with x50 magnification. Printed in surface option surface mode (two lines) (A). Printed in normal option surface mode (single line) (B). Source: self-made. 

This phenomenon could be because on normal option, the extruder sometimes don’t print consecutively (printing the sections in Random form (Figure 5), which does not happen with the surface option.

Figure 5 Extruder path in surface mode normal option. 

The disadvantage of using two lines to form one is a thicker line results, so, the porosity of the scaffold decreases. To compensate part of this effect, the line is designed 0.01 mm smaller for this geometry.

Dimensions and porosity for PLA

Geometry

On average, all the geometries except the ellipse were approximately 50 µm smaller than the computationally modeled design. For the pore size of the ellipse to agree with that of the other geometries, an overlap had to be designed as shown Figure 6. A value of 100 µm was established to compensate for part of the material shrinkage.

Figure 6 Dimensions in microns of ellipsoidal scaffolding concordant with pore. Source: self-made. 

Porosity

Table 8 shows the porosity percentage achieved for each PLA scaffold.

Table 8 Porosities of PLA scaffolds. 

	Square	Triangle	Ellipse
𝒑1	86%	87%	60%
𝒑2	80%	82%	56%
𝒑3	73%	78%	-
𝒑1	86%	87%	60%

Finally, Figure 7 shows a magnified detail of the PLA studied geometries.

Figure 7 Detail of the geometries printed in PLA. Image from an optical microscope with 50x magnification. Square scaffold (A). Triangular scaffold (B). Ellipsoidal scaffold (C). Scale bars = 500 µm. Source: self-made. 

Manufacturing observations for PCL

For the printing of the PCL scaffolds, it was necessary to install a hot plate on the base of the printer, to avoid the adhesion problems to the base for the rapid cooling suffered by the material. This phenomenon causes a contract and forms small drops that harden and modify the height printing. When this occurs, the needle passes through the sector again, it hits the drops, and the scaffolding comes off. Together with the previous point, other conditions to ensure surface adhesion, is the use of rough tape on the hot bed.

The pore width should be maintained between 300 and 600 µm since at the lower limit the filaments are joined and at the upper limit they are cut (Figure 8).

Figure 8 PCL scaffolding cut filaments (A), agglomerated filaments (B). The images were taken from optical microscope with x50 magnification. Source: self-made. 

Dimensions and porosity for PCL

Geometry

Due to the characteristics and temperatures at which the PCL was printed, it was deposited in a liquid state, and as a result, the material spreads occupying a larger surface compared to the computationally modeled design. This results in the pores that were designed to 400 µm being an average of 250 µm.

Porosity

Based on the above and considering the largest diameter of the extruder, the porosity was reduced to 34% for the square geometry. Figure 9 (A) shows a magnification of the final pore size. In Figure 10 it can see a 20 mm PLC scaffold.

Figure 9 PLC Scaffold. Detail of the geometries printed in PLC. Image from an optical microscope with 50x magnification. Scale bars = 500 µm (A). 20 mm PLC scaffold (B). Source: self-made. 

CONCLUSIONS

The three-dimensional elaboration of scaffolding for the regeneration of cartilaginous tissue represents a challenge for the existing manufacturing market devices. From this study, it is concluded:

For the equipment used in this work, the pore size of PLA must be greater than 200 µm to prevent the effect called "elephant foot" enclosing the lower pores, while for manufacturing in PCL, the size of the computationally model pore should be greater than 300 µm to prevent the filament joining together and less than 600 µm, as the largest size, because the filaments will be cut.

It is necessary to consider the geometric design, the contraction of the geometries at least 50 µm for the PLA.

Particularly, for the ellipsoidal scaffolds of PLA, an overlap had to be designed to compensate the material retraction.

In PCL material, this difference is intensified, therefore, it should be considered that the computationally designed pore, decreases by approximately 150 µm in the final printed scaffold.

The extruder temperature is one of the most important variables. In PLA material, printing can be affected when the ambient temperature rises to 20 °C, the extruder temperature must be reduced below the manufacturer temperature indicates.

This is extremely important when printing is a very low speed. In PLC case, this is even more sensitive to temperatures, because it is a material that undergoes large retractions if it cools quickly, so it is necessary to implement a heat bed to the printing surface. On the other hand, the high temperature remains in a liquid state affecting the pore size.

As recommendations and future research work it is suggested: