基于对T2 mapping的研究，首次发现静置法移植脂肪水分含量在术后1年内从高到低变化，在一年时候最接近正常脂肪。

静置法获取的移植脂肪，术后其水分含量变化从高到低，且其含量与移植脂肪的体积保留率成正比，在术后3个月，移植脂肪含水量仍然高于正常脂肪，提示移植脂肪仍然存在水肿，在术后1年移植脂肪含水量与正常脂肪无差异。

脂肪移植术后3个月内，因移植脂肪人眼不可见的水肿存在，提示以往研究方法算出的移植脂肪保留率，都高于真实体积保留率。

Background

Aging, trauma, and disease can result in soft tissue defects and affect the appearance, which can be improved by the use of a variety of filler materials. Autologous fat transplantation is a method of redistributing autologous fat to improve the appearance of patients with soft tissue defects. As one of the filling materials, autologous fat has good histocompatibility, seldomly leads to rejection and allergic reaction, and it is widely used for facial defects, facial rejuvenation, breast filling, hip filling, and so on. The researchers used imaging methods to measure the rate of volume retention after autologous fat transplantation in order to evaluate the surgery outcome and improve the surgery procedure. Current evaluation methods include ultrasound, CT, 3D facial scanners, and magnetic resonance imaging, or even subjective evaluations by patients or doctors. However, the inaccuracy or insensitivity of these methods limits their application to the small size of the fat volume. Ultrasound imaging, for example, inevitably puts pressure on soft tissue to change its shape; CT imaging has low sensitivity of adipose tissue, which is unable to distinguish fat between graft and native fat. The ionizing radiation of CT's x-rays inhibits its widespread use. In addition, the fat-suppressed MRI sequence often used in the field of plastic surgery is also difficult to delineate the fat profile. The mechanism of the MDIXON sequence is based on chemical shift effects, which separates water and fat signals, producing In-phase images and water-only images (fat suppression images) and fat-only images. Water-only images are often used in clinical diagnosis of disease, but fat-only images are rarely used alone. In addition to fat quantification, pure fat images are usually discarded, but they provide high-quality anatomical images of fat tissue without the need for contrast media or image post-processing.

Objective

We hypothesized that the structure of the newly transplanted fat after centrifugation, filtration and other processing is different, and it can be reflected from the image. The fat-only image provides a visualization of the transplanted fat, helping to distinguish the autogenous fat from the original fat. The purpose of this study was to evaluate the application of Dixon fat-only images in the imaging of autogenous fat grafts, to study the signal and volume changes of transplanted fat during the recovery of adipose tissue within one year after autogenous fat transplantation, and to evaluate the feasibility of fat-only imaging with autologous fat graft.

Methods

Four volunteers were recruited to optimize the Magnetic Resonance Imagin protocol at the start of our study. Between August 2019 and October 2020, a total of 68 patients with facial defects were enrolled in the study. All the patients were randomly selected for MRI examination 7 days, 3 months or 1 year after autologous fat transplantation. Of these, 21 patients canceled surgery due to COVID-19 isolation, and 13 were excluded from the MRI because of geographic reasons and travel restrictions for the new coronavirus disease in 2019, 11 patients were removed because they received injections of hyaluronic acid or autologous fat prior to the study. Eventually, 23 patients were prospectively enrolled in the study.

1. Surgical procedure: the procedure consists of three steps: liposuction, fat preparation and fat filling. Depending on the patient's choice, autologous fat is removed from the abdomen or thighs. The swelling fluid containing 400 mg of lidocaine and 1 mg of epinephrine per 1000 ml of saline was infused into the subcutaneous layer. Under local anesthesia, liposuction was performed with a 2.0 mm sharp cannula connected to a 20 ml hypobaric syringe. The cannula was moved back and forth to collect the autologous fat, and the fat collected was deposited for about 15 minutes. During the precipitation process, the fat cells were separated from the serum and oil and transferred to a 1 ml syringe for injection.

Two holes of about 0.5 mm were punctured on each patient's bilateral zygomatic bone or nasal base with an 18-gauge needle. A 1 ml syringe containing fat was connected to an injection cannula and the well-prepared fat was injected into the deep fat layer. Intraoperative volume was recorded for each patient at each site, and postoperative outcome was assessed by both the patient and the doctor, as excellent, good, average or poor.

2. MRI facial scan: The patient was asked to lie down in a 32-channel head coil to receive MRI facial scan. Scan range: up to the top of the forehead, down to the chin, outside to the ears. The magnetic resonance imaging uses a 3.0 T unit with a 32-channel head coil (Philips HealthTech) to assess the patient's facial fat signal, shape and volume. The time of examination was 7 days, 3 months or 1 year after the operation. Image evaluation is performed by both a radiologist and a plastic surgeon, and the image is outlined and scored using a professional scale system. The signal contrast between the transplanted fat and the injected fat, the volume of the transplanted fat and the T2 mapping value of the transplanted fat were measured.

Results

A total of 23 women participated in surgical and magnetic resonance imaging evaluation, and 46 facial images were obtained. The operative sites included 26 apple muscles and 20 nasolabial sulcus. 8 patients were followed up on 7 days, 9 on 3 months, and 6 on 1 year after surgery. There was no significant age difference among the three groups (p = 0.2). The average volume of transplanted fat was 2.60 ± 0.53 ml. Postoperative satisfaction: Excellent 19 cases (82.6%), good 2 cases (8.7%), General 1 case (4.3%), poor 1 case (4.3%), a total of 21 cases (91.3%) patients (excellent N = 19, good n = 2) were satisfied with the results. The result of operation was evaluated by doctors: excellent 18 cases (78.2%), good 4 cases (17.3%), General 1 case (4.3%). So far, there have been no complications and no surgical site infections. 3 patients received a second injection. 

For mDIXON imaging evaluation, the median of T1W image quality and differentiation is 5 and 5, respectively. The median values of T2W image quality and differentiation were 4 and 3 respectively, the difference of T1W and T2W image quality was significant (p < 0.01), and the difference of T1W and T2W image differentiation was also significant (p < 0.0001). MRI showed complete absorption of injected fat on the left side in 1 case 1 year after surgery. Therefore, only 45 sides of signal contrast were recorded. After 7 days, 3 months and 1 year, the signal contrast ratio of transplanted fat was 28.8 ± 4.7%, 16.33 ± 2.1%, 3.3 ± 1.3%. There was significantly difference in signal contrast between 7 days and 3 months postoperatively (p < 0.0001), and between 3 months and 1 year postoperatively (p < 0.0001).

Pearson correlation analysis was used to evaluate the relationship between the amount of injected fat and the amount of transplanted fat on the 7th day. There was a positive correlation between the two variables (r = 0.9556; p < 0.0001). The average measured volume was 2.6 ± 0.47 ml, on the 7th day after surgery. The average volume retention rate of injected fat was 94.1 ± 5.75%, the fat volume retention rates at 3 months and 1 year postoperatively were 48.7 ± 17.34% and 33.1 ± 12.94%, respectively. The volume retention rate was significantly different between the 7th day and 3rd month, 3rd month and 1st year(p < 0.001).

The mean T2 mapping values of normal fat, 7 days, 3 months and 1 year postoperatively fat were 137.4 ± 6.215 ms, 172.9 ± 12.78 ms, 154.4 ± 11.14 ms and 135.9 ± 9.275 ms. At the 7th day after surgery, the T2 mapping values of the transplanted fat and the normal fat were significantly different (p < 0.0001), at the 3rd month after surgery, the T2 mapping values of the transplanted fat and the normal fat were significantly different (p < 0.0001), and at the 1st year after surgery, the difference between transplanted fat and normal fat disappeared (p = 0.70). There were significant differences in T2 mapping between 7 days and 3 months, 3 months and 1 year postoperatively (p < 0.0001).

According to Pearson's Linear independence analysis, it was found that, there was no linear relationship between T2 mapping and Signal Contrast at 7 days, 3 months and 1 year postoperatively (7 days: R = 0.09, P = 0.73; 3 months: R = 0.28, P = 0.26; 1 year: R =-0.08, P = 0.81).

Conclusion

1. This study was the first time to use the signal contrast and T2 mapping sequence of MRI to directly detect the changes of the edema of autologous facial fat in human body.

2. In the magnetic resonance imaging (MRI) method of facial autologous fat transplantation, T1W fat-only images obtained by Dixon imaging showed that the grafted fat was more effective than T2W, and had better image quality and resolution, helping doctors draw a clear picture of the transplanted fat.

3. Based on fat-only images, we demonstrated for the first time that the T1W signal of grafted fat was significantly different from that of facial fat during the 1-year recovery of autologous fat transplantation.

4. During the 1-year recovery period after fat grafted, the signal contrast of grafted fat gradually decreased, and it turned the closest to normal fat at the end of the year.

5. During the 1-year recovery process after fat grafted, the volume of fat gradually decreased and the absorption rate gradually slowed down.

6. It was first time to find that there was a significant difference between the T2 mapping value of the grafted fat and that of the facial fat during the 1-year recovery period.

7. Based on the T2 mapping study, it was first time to find that the moisture content of the grafted fat changed from high to low within 1 year after the operation, and was the closest to the normal fat at post-op 1 year.

8. The moisture content of the grafted fat changed from high to low after operation, and its content was proportional to the volume retention rate of the grafted fat. At 3 months after surgery, the moisture content of the grafted fat was still higher than normal fat, the results showed that there was still edema in the grafted fat, and there was no difference in moisture content between the grafted fat and the normal fat 1 year after surgery.

9. Within 3 months after autologous fat transplantation, there was no visible edema in the eyes, which suggested that the fat retention rate calculated by previous research methods was higher than the true volume retention rate.