Choosing the Optimal Tissue Expander Size: A Single Center Randomized Clinical Trial on Chinese Microtia Patients’ Normal Ears Using 3D Scanning


 The tissue expansion technique is the most suitable procedure for Chinese patients with microtia. However, it is difficult to determine whether the expanded flap is sufficient, and there are no clear or objective guidelines for determining the volume of the expander for different sizes of auricles. One hundred patients with unilateral microtia who visited our department in 2021 were randomly selected for auricular data collection using 3D scanning. The auricle length, width, projection, perimeter, and surface area were measured. Eight different volumes of expanders underwent CT and the surface areas of these expanders were measured. The surface areas of the auricles and expanders were compared and the correlation between them was explored. The average auricle parameters were calculated. The scatter plot showed a linear relationship between auricle length and auricle surface area (R2 = 0.9913), which demonstrated that the auricle area was approximately equal to the auricle length multiplied by 76.921. Additionally, the surface area of the expanders was measured and made into a table for selection against the surface area of the auricles. Using our equation, the auricle surface could be estimated by simply measuring the non-defective auricle length; therefore, the suitable volume of the expander could be determined.


Introduction
For unilateral microtia, the ear on the non-defective side is of vital importance. In external ear reconstruction, plastic surgeons must perform the surgery in reference to the normal ear so that the organs on both sides are symmetrical. The more similar the two ears look, the more successful the surgery is. Thus, quantifying the parameters of the auricle has become a decisive factor in auricular reconstruction. There are many innovative ways to measure the auricle [1][2][3] . However, few of these are considered precise, objective, economical, time-saving, and easy to cooperate with children, who are the main group of patients with microtia who visit the hospital. Three-dimensional (3D) scanning has advantages in solving the problems mentioned above [4][5][6][7] and has now become a routine preoperative procedure in our department for auricular data collection. The scanning device is mobile and portable. The data scanned are objective and accurate, and the subsequent data analysis is not constrained by time and location.
Tissue expanders also play an important role in auricular reconstructive surgery. In the 1950s, Neumann rst used a tissue expander for auricular reconstruction 8 . Expansion-method auricular reconstruction was introduced into China in the 20th century and has become the mainstream method ever since then 9,10 . It is considered that Asians have tighter skin in the mastoid region behind the ears; thus, the expansion method is more suitable for the Chinese because it can provide a more adequate ap area for a manmade auricle 11 . Tissue expanders are extremely elastic silicone balloons and are always over lled to generate enough tension to expand the skin. Therefore, the nal volume of an expander is highly dependent on subjective decisions rather than the speci cation volume. To date, however, there is no clear reference guide on how to determine the expander's volume for different auricles. Surgeons rely on subjective feelings to decide what speci cation of expander to choose and the ll volume. The relationship between the expanded ap and auricle size remains vague, and it is di cult for surgeons to judge whether it is su cient and safe to use the expanded ap to completely wrap the auricle framework.
The purpose of this study was to measure the normal ear parameters of patients with microtia among the Chinese population using 3D scanning, explore an easier way to estimate the auricle's surface area, clarify the theoretical expansion area of different volumes of expanders, try to nd the simplest correspondence between expander size and auricle size, and develop a reference guide for the selection of expanders' speci cation and injection volume in auricular reconstruction.

Results
One hundred patients (85 men and 15 women) with unilateral microtia (29 on the left side and 71 on the right side) were selected for this study. The mean age of the patients was 9.75 ± 4.2 years. The average auricle length was 58.88 ± 4.77 mm, the average auricle width was 33.15 ± 2.75 mm, the average auricle projection was 20.92 ± 2.70 mm, the average auricle perimeter was 104.67 ± 8.40 mm, and the average auricle surface area was 4502.1479 ± 633.4493 mm 2 . The data are presented in Table 1. A scatter plot demonstrating the relationship between auricle length and surface area was drawn automatically using Excel, as shown in Fig. 1. After excluding one set of data with signi cantly larger deviations (the 2nd set), a linear trend line of the scatter plot was generated. When setting the intercept as 0, the equation of the trendline is where y is the auricle surface area and x is the auricle length (R 2 = 0.9913).
The surface areas of expanders of different sizes are listed in Table 2. Half of the area of these expanders was also measured to represent the effective expansion area. As the injection volume increased, the expander surface area increased almost linearly (Fig. 2). The 50-, 60-, and 70-ml volumes were measured using a 50-ml speci cation expander. The remaining volumes were measured using a 100-ml speci cation expander.

Discussion
Auricular reconstruction is a challenging procedure. A normal external auricle is a complex structure that has a certain degree of hardness and elasticity, with many 3D details. In 1959, Tanzer introduced the autologous rib cartilage technique 12 . Since then, the best material for auricular reconstruction has always been the costal cartilage; Brent and Nagata, and many other scholars, later modi ed Tanzer's technique and developed their own methods [13][14][15][16][17] . To wrap the auricle framework carved with the costal cartilage, the ap amount should be su cient. Otherwise, the tension of the ap will be too large, especially where the pressure is highest in the upper part of the helix, causing gradual ischemia, rupture of the ap, and eventual exposure of the cartilage [18][19][20] . However, it is di cult for surgeons to determine whether the ap is adequate. Undeniably, the area of the skin ap must be at least equal to the surface area of the contralateral auricle (non-defective side), so that the ap can exactly cover the framework. Therefore, anthropometric assessment of the normal ear is key to quantifying the surgical needs.
Many studies have attempted auricle anthropometry, but the methods vary widely. Most anthropometric methods have always been manual measurements, in which the data are subjective and inaccurate, and post-processing cannot be performed. In addition, the surface area of the auricle could not be accurately measured. Although CT before surgery can also achieve this goal, the cost is high, and patients must be exposed to unnecessary radiation. The 3D scanning method used in this study is noninvasive, convenient, portable, economical, and less time-consuming. In addition, since most patients with microtia are children, the non-noisy scanning method is easier for them to cooperate with.
The application of expanders in auricular reconstruction is complicated and has several controversial issues. The most important issue is the quantitative relationship between the expander and the expanded ap. Some studies have scanned the expanded skin surface area using an expander under the skin 21 . However, in most cases, expanders with the same speci cation and volume can provide different expanded skin areas. There are two main reasons for this nding. First, the patients had different skin and soft tissue thicknesses. Patients with thicker dermis and subcutaneous fat are more di cult to expand because the tension generated by an expander of the same size is smaller in this case, and a smaller area of the expanded skin will be acquired. Second, microtia is often accompanied by varying degrees of hemifacial microsomia, which manifests as different degrees of depression of the skull in the mastoid area compared to the contralateral side. In this case, the position where the expander is buried under the skin will be relatively deep, and the part protruding from the skin level will be small; hence, the effective expansion area will be smaller. Therefore, in this study, we measured the full surface area of expanders of different sizes in vitro to prevent the in uence of these factors on the results. The effective expansion area of the expanders was uniformly considered as half of the full area of the expanders. Our study showed that as the amount of saline injected increased, the expansion area increased, and the two were almost linear. Two speci cations of kidney-shaped expanders (50 and 100 ml) are most commonly used in clinical practice. In most cases, both types of expanders can generate a su cient number of aps ( Fig. 3). However, so far, there is no uniform standard for how to choose these two expanders, and opinions vary among surgeons.
To facilitate the selection of expanders in auricular reconstruction surgery, we constructed a scatter plot of the measured patient's auricle length and auricle area to clarify whether there was a connection between the two. The trend line of the scatter plot was automatically generated using Excel. We found that the correlation coe cient (R 2 value) was closest to 1 only when we chose the trendline as a linear one and set the intercept to 0, which means that there is a strong linear relationship between auricle length and area (R 2 = 0.9913). Using Equation (a), we can roughly estimate the surface area of the auricle by simply entering the auricle length. By comparing the results to the surface area of different expanders (Table 2), we can clarify which dilator should be selected to produce a su cient area to completely wrap the auricle framework. However, considering that the elastic shrinkage of the expanded skin and the ap tension cannot be too large after surgery, a slightly larger (at least 10 ml larger) expander should be selected.
Combining our results with our surgical experience, we recommend that it is safe and has a satisfactory surgical outcome of completely wrapping the auricular framework with an expanded ap for patients whose auricle length is less than 58 mm. In this case, the auricular surface area is approximately 4461.42 mm 2 by equation (a), and a 50-ml expander over lled with 70 ml of saline could be selected according to Table 2. For patients with auricle lengths greater than 58 mm, we do not recommend completely wrapping the auricle framework with an expanded skin ap. An oversized framework causes greater pressure on the skin ap at the helix, leading to avascular necrosis and rupture of the skin ap, followed by an exposed framework. In addition, the hairless skin on the affected side of the microtia is limited, and an excessively large expanded skin ap will inevitably expand the scalp too much, causing the reconstructed auricle to have too much hair and affect its appearance. For patients with auricle lengths greater than 58 mm, we recommend wrapping the front of the auricular framework and most of the helix with an expanded skin ap and then lifting the temporal muscle fascia ap behind the framework and grafting skin on its surface (Fig). This method not only relieved the tension of the skin ap but also made a higher auricle and a clearer and more apparent ear-cranial groove (Table 3). In conclusion, 3D scanning is non-invasive, easier for children to cooperate with, more accurate for data measurement, and less time-consuming for auricular data collection. The auricle length and surface area had a roughly linear relationship. The equation is as follows: where y is the auricle surface area and x is the auricle length (R 2 = 0.9913). Simply measuring the length of the auricle could estimate the area of the auricle and guide the choice of the expander volume.

Methods
This study was reported in accordance with the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) statement 22 .
Ethics approval and consent to participate Patients and their families provided written informed consent for participation in the study. The patients/legal guardians agreed to use their images for publication of this article and signed the informed consent. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This study was approved by the Medical Ethics Committee of Plastic Surgery Hospital, Chinese Academy of Medical Sciences ( le no. 2020-186).

Participants
For auricle anthropometry, 100 patients with microtia who visited our department in 2021 were randomly selected for the study before auricular reconstruction surgery. All participants in this study were patients with unilateral microtia patients aged 5-10 years. Other exclusion criteria were as follows: 1) patients with any auricular deformity of the healthy side, such as prominent ears, Stahl's ears, and cryptotia; 2) patients with severe hemifacial microsomia; 3) patients with auricular trauma and/or surgery history of the healthy side; and 4) extremely uncooperative patients.

Measurements
Auricle anthropometry Patients were asked to remove all types of earrings or headwear and have a haircut to fully expose the bilateral ears. The heads of all selected patients (including the entire face and bilateral ears) were scanned using an Artec Space Spider 3D Scanner (Artec 3D, Redmond, USA) (Fig 4). The scanning data were synthesized into STL les using Artec Studio 10 (Artec 3D, Redmond, USA) and then imported into 3matic Research 9.0 (Materialise NV, Leuven, Belgium), where all the measurements were performed ( Fig  5). The following ve factors of the normal ear of the enrolled microtia patients were measured (Fig 6). Ear surface area -Surface area of the whole auricle, measured from the perpendicular line of the helix foot and tragus notch to the ear-cranial groove All the data were collected and entered into a table. The length and surface area of the auricles were plotted on a scatter plot, and a linear regression equation was calculated.

Expander volume measurement
For expander volume measurement, two speci cations (50 ml and 100 ml) of kidney-shaped soft tissue expanders (Jiusheng Medical Supply, Yuyao, China), which are most commonly applied in expansion auricular reconstructive surgery, were used in this study (Fig 7).
Three 50 ml expanders were injected with 50 ml, 60 ml, and 70 ml of saline. Five 100 ml expanders were injected with 80, 90, 100, 110, and 120 ml of saline. All eight expanders underwent computed tomography (CT) after injection (Brilliance CT 64 slice, Philips Medical Systems, Cleveland, OH; tube voltage, 120 kVp; tube current, 220 mAs; collimation, 0.6 mm; pitch, 0.8; rotation time, 0.75 s; matrix, 512 512; and eld of view, 350 mm). DICOM data were then acquired and imported to ProPlan CMF 3.0 (Materialise NV, Leuven, Belgium), where the injection hose and injection pots were removed manually and STL les of expanders were created. All STL les were then imported into Geomagic Wrap 2015 (3D Systems Inc., Rock Hill, USA), and the surface area of the expanders was measured automatically using the software (Fig 8).

Statistical analysis
All anthropometric data were subjected to statistical analysis using SPSS 25 (IBM Corp., NY, USA). The trend line and linear equations were performed using Microsoft 365 Excel. Figure 1 The scatter plot, linear trendline, and equation that demonstrate the relationship between auricle length and surface area. R 2 , coe cient of determination.

Figure 2
Relationship between expander's injected volume and expander's half surface area.

Figure 3
The speci cations of the two expanders that are used in our clinical work. Both expanders can obtain satisfactory aps, but there are no clear guidelines for their selection and application.

Figure 4
Method and devices of auricle scanning.

Figure 5
Scanning data were collected and synchronized into STL les by Artec Studio 10.   The speci cations of the two tissue expanders that are most commonly used in auricular reconstruction. Left 100 ml, right 50 ml. Figure 8 DICOM data of the expanders were processed in ProPlan (above), and the surface area was measured in Geomagic Wrap (below).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. SupplementaryInformation.docx