A New Ex Vivo Model for the Evaluation of Endoscopic Submucosal Injection Material Performance

Both endoscopic submucosal dissection (ESD) and endoscopic mucosal resection (EMR) are currently common treatments for early-stage gastrointestinal cancer^1,2. Injecting a submucosal injection material (SIM) into the submucosa is one of the most important steps for both the EMR and ESD procedures^2,3. High submucosal elevation and maintenance of submucosal elevation are critical criteria for safely conducting EMR/ESD.

Although normal saline (NS) has been used as a SIM since the invention of endoscopic therapy^4,5, sodium hyaluronate (HA) was introduced as a treatment in recent years^6,7. HA became widely used in endoscopic treatments as a superior SIM due to its high performance^8,9,10,11. Currently, a performance comparison between the existing SIMs was conducted, and high-performance SIMs were developed to identify another superior SIM^{5,12,13,14,15,16,17,18}.

The ex vivo model using a porcine stomach specimen has been used to evaluate SIM performance, because the estimation of SIM performance in the human gastrointestinal tract is very difficult^19,20,21,22. However, this conventional ex vivo model is extremely simple, and has the scope for improvement. Reproducing an environment closer to the human gastrointestinal mucosa will enable accurate evaluation of SIM performance.

In our previous study, we developed a new ex vivo model that can be used to evaluate the performance of various SIMs in detail by applying constant tension to the specimen's ends. We also confirmed that the proposed new ex vivo model, allows accurate SHE measurement under uniform conditions and a detailed comparison of the performances of various types of SIMs²³.

In this study, we present a complete appearance of the new ex vivo model, and the detailed setup methodology of the new ex vivo model is explained in detail using videos and figures. The new ex vivo model consists of parts that are easily available and can be quickly set up. Descriptions of detailed setup methodology will contribute to the dissemination of the new model.

The following protocol follows the animal care guidelines of the Kyoto Prefectural University of Medicine.

1. Preparation of Specimens Using a Porcine Stomach

NOTE: The first step is to prepare specimens to be used in the ex vivo model (Figure 1). The thickness of the porcine gastric wall varies in different areas of the stomach. Use the upper third of the porcine stomach, which is relatively similar to the human stomach. Exclude inappropriate specimens where submucosal elevation is not found due to fibrosis.

1. Cut gastric specimens into squares with approximate dimensions of 6 × 6 cm.
2. Store the gastric specimens immediately at a temperature of -30 °C.
3. Thaw frozen gastric specimens right before the measurement procedure to ensure uniform measurement conditions.

2. Detailed Setup Methodology of a New Ex Vivo Model

NOTE: Stretch out the thawed specimen on a board in two different ways. In the conventional ex vivo model, fix the specimen with pins (Figure 1A)^19,20,21,22. On the other hand, in the new ex vivo model, fix or stretch both ends of the specimen with clips to produce a constant tension (Figure 1B, C). All parts of the new model are easy to obtain, and the setup of the new model can be completed quickly (Figure 2). The procedure of the new model is as follows (Figure 3):

1. Connect the stainless-steel clip and the key wire and the S shaped hook (Figure 3A).
2. Connect the wire and the S shaped hook and the weight (Figure 3A).
3. Connect the hook to the other end of the wire. A traction device is completed in the above process (Figure 3B).
4. Fix the pulleys (Figure 2b) at both ends of the base (Figure 3C).
5. Place the rubber plate (6 x 6 cm) on the center of the base (Figure 3C).
6. Place the gastric specimen on the rubber plate and pinch the specimen ends with the clip of the traction device.
7. Hang the weight through the pulley (both side). Thereby, constant tension can be applied to the specimen (Figure 4).
8. Start the measurement of SEH, because after the setup of the new model is completely finished (See Step 3 below).

3. Evaluation of SIM Performance

NOTE: In this study, we used normal saline (NS) and 0.4% sodium hyaluronate (HA) as SIMs to be tested, and measure SEH of the two SIMs. Three independent measurements are performed. The obtained data are expressed as the mean and standard deviation (S.D.). Statistical analysis was performed by using the statistical analysis software (GraphPad Prism 7). We analyzed continuous variables (SEH) with the Student's t-test, and the magnitudes with P < 0.05 were considered significant. The measurement of SEH is as follows (Figure 5).

1. Perform zero-point adjustment of the height gauge, based on the height of mucosa before a submucosal elevation procedure. In detail, perform zero-point adjustment by pushing the PRESET button after fixing the scriber at the height of the mucosal surface.
2. Inject 2.0 mL of each solution horizontally into the submucosa from the specimen margins using a 2.5-mL syringe and 23-gauge needle, to perform a submucosal elevation procedure (Figure 5A-C).
3. Measure SEH promptly using a digital height gage at 0, 2.5, 5, 7.5, 10, 12.5, 15, 17.5, 20, 30, 45, and 60 min after the injection (Figure 5D). In detail, record the height displayed on the height gauge when fixing the scriber to the top of the submucosal elevation.
4. Perform three independent measurements, and express the obtained results as the mean and standard deviation.
5. Analyze the obtained data using appropriate statistical software and evaluate the performance of SIMs (The performance can be compared between each SIM.)

SEH was measured over time in the new ex vivo model or conventional ex vivo model. The values of SEH (NS) measured using the conventional model [NS was injected into the submucosa of the specimen fixed with pins (0.0 N)] were 5.7 mm (0 min), 3.6 mm (5 min), 3.0 mm (10 min), and 2.2 mm (30 min).In this way, the values of SEH decreased with increasing post injection time.A similar analysis was performed using 0.4% HA instead of NS. The values of SEH (0.4% HA) were 6.5 mm (0 min), 5.2 mm (5 min), 4.8 mm (10 min), and 4.1 mm (30 min).The resulting SEHs of 0.4% HA were higher than those of NS regardless of the post injection time.The SEHs (NS and 0.4% HA) obtained using the conventional model (in the absence of the applied tension) exhibited relatively large variations (in other words, their standard deviations were high) (Figure 6A).

Next, the values of SEH (NS) measured using the conventional model [NS was injected into the submucosa of the specimen stretched at a constant tension (1.5 N)] were 4.8 mm (0 min), 3.0 mm (5 min), 2.4 mm (10 min), and 1.8 mm (30 min). When the tension was increased to 3.0 N under the same conditions, the values of SEH (NS) were 4.5 mm (0 min), 2.3 mm (5 min), 1.5 mm (10 min), and 1.3 mm (30 min).The SEH measured at various post injection times decreased with increasing tension. The SEHs obtained using the new model exhibited small variations (in other words, their standard deviations were low) (Figure 6B, C).

For evaluating the relationship between SEH and tension applied to the specimen, we compared SEH measured at different tensions (0.0-3.0 N). In the analysis with the new model, the SEH obtained at a tension of 3.0 N was significantly lower than the SEH obtained at a tension of 1.5 N (in all cases, the condition P < 0.001 was satisfied). In contrast, since standard deviations of SEHs obtained using the conventional model (0.0 N) were high, there was no significant difference between SEHs obtained using the conventional model (0.0 N) and the new model (1.5 N) (Figure 6D, E).

Figure 1. New ex vivo model and conventional ex vivo model. In the conventional ex vivo model, the porcine specimen was fixed with pins (A). On the other hand, in the new ex vivo model, both ends of the specimen were stretched with clips to produce a constant tension (B). This model can be tensioned uniformly by using a weight, and the tension can be arranged by changing the weight (C). Each SIM was injected into the submucosa of the specimen, leading to submucosal elevation (D). This figure has been modified from Hirose et al.²³. Please click here to view a larger version of this figure.

Figure 2. All parts used for the new model. The new ex vivo model consists of parts that are easily available. All parts used for the new ex vivo model: (a) approximately 50-300 g of weights (the weight can be changed appropriately depending on the applied tension); (b) fixed type pulley with pulley diameter of 25 mm; (c) stainless steel wire with a diameter of 0.45 mm; (d) stainless steel clip of width 147 mm; (e) stainless steel key wire with a length of 12 cm; (f) stainless steel S shaped hook; (g) lockable stainless steel S-shaped hook. (This figure has been modified from Hirose et al.²³). Please click here to view a larger version of this figure.

Figure 3. The detailed setup method of the new ex vivo model. The new ex vivo model can be quickly set up. (A) Connect the stainless steel clip (Figure 2d) and the key wire (Figure 2e) and the S shaped hook (Figure 2g). Next, connect the wire (Figure 2c), the S shaped hook (Figure 2f) and the weight (Figure 2a). (B) Finally, connect the hook (Figure 2g) to the other end of the wire (Figure 2c). A traction device is completed in the above process. (C) Fix the pulleys (Figure 2b) at both ends of the base [rectangular wooden base (45 x 60 cm) for assembling the model]. Next, place the rubber plate (6 x 6 cm) on the center of the base. Please click here to view a larger version of this figure.

Figure 4. The complete appearance of the new ex vivo model. Accurate measurement of SEH can be performed. Please click here to view a larger version of this figure.

Figure 5. The measurement procedure by using the new ex vivo model. To evaluate SIM performance, the magnitude of SEH was measured by a digital height gage (A). Using a 2.5-mL syringe with a 23-gauge needle, 2.0 mL of each SIM was injected into the submucosa from the specimen margins to create a submucosal elevation (B, C). The digital height gage was used to measure of the height of the submucosal elevation (i.e., the values of SEH) (D). Please click here to view a larger version of this figure.

Figure 6. Measurement of SEH using either the new or conventional model. After the injection of NS or 0.4% HA into the submucosa of the specimen fixed with pins (0.0 N) (A) or stretched at a constant tension (1.5 N or 3.0 N) (B, C), SEH was measured using the height gage. Next, we compared the values of SEH measured at different tensions (0.0, 1.5, and 3.0 N) after the submucosal injection of NS (D) or 0.4% HA (E). Data are expressed as mean ± S.D. of more than three independent experiments. (This figure has been modified from Hirose et al.²³) Please click here to view a larger version of this figure.

The porcine stomach used for the new model should be stored in a freezer immediately after resection, and be used within a few months after freezing, since the freshness of the swine stomach is essential for SEH measurement. (We measured SEH using both frozen and unfrozen gastric specimens, and confirmed that there was no difference in the result of SEH measurement.)

The quality of gastric specimens is greatly influenced by the individual differences of porcine stomachs. Hence, it is recommended to exclude obviously thick specimens or specimens with many folds before measurement. Furthermore, some specimens may be inappropriate specimens for SEH measurement due to fibrosis. It is recommended to exclude the inappropriate specimens where submucosal elevation is not found due to fibrosis.

Since the digestive tract is expanded by endoscopic treatment, some tension is applied to the gastrointestinal mucosa. It was revealed that SIM performance (evaluated by measuring the values of SEH) decreased with increasing the values of tension applied to the specimens. Therefore, the tension was an important factor affecting the SIM performance (i.e., the values of SEH)²³. The application of tension of 1.5-3.0 N can reproduce an environment closer to the human gastrointestinal mucosa. However, a limitation of this method is that the optimal tension may depend on the difference of the specimen used for analysis.

In the conventional model, since the tension applied to each specimen varies depending on the degree of specimen fixation, the variations of measured SEH are large (which correspond to the high standard deviations of SEH). Therefore, these high standard deviations make it difficult to compare each SEH in detail and perform statistical analysis. On the other hand, owing to small variations of SEH measured in the new model, SIM performance can be compared accurately ex vivo and precise statistical analysis is performed.

In conclusion, the new ex vivo model enables accurate SEH measurement and detailed comparison of SIM performance. Descriptions of detailed setup methodology will contribute to the dissemination of the new model and the development of high-performance materials.