内镜辅助面部年轻化两种入路（含深部 SMAS 提升）的安全性与满意度纵向研究

一分钟要点

• 研究背景：除皱术从单层皮瓣演进到针对 SMAS（表浅肌筋膜系统）等深层结构，效果更自然持久；但经耳前大切口仍有面神经损伤与皮瓣坏死风险。本研究评估两种内镜入路（含深部 SMAS 提升）的并发症、美学结果与患者满意度。
• 两种术式：① 内镜上 2/3 面部提升（发际 5 mm 内镜、5 个切口，沿 SMAS 同层下分离至中面部，PDS 3-0 将 SMAS 折叠悬吊至颞筋膜，联合上下睑成形）；② 全脸提升 = 内镜上 2/3 + 开放下脸（开放颈阔肌下/SMAS 解剖与折叠、必要时颏下入路处理颈阔肌/二腹肌/颌下腺）。提升主体均为深部 SMAS 解剖与折叠。
• 样本与随访：加拿大单中心混合纵向观察（2021-01 至 2023-12），35 例均为女性，中位年龄 58 岁（41–76）；约半数分入两组。随访 1/3/6/12 月，用 House–Brackmann 测运动神经、改良感觉测试测感觉神经、FACE-Q 测满意度。
• 安全性：术后 1 月运动神经障碍高达 82.8%（颞支最常见），但 12 月内全部恢复；感觉障碍除 1 例滑车上神经轻度持续感觉减退外均缓解；全脸组 1 月颊支受累更多（50%，P=0.0003）。其他副作用（紧绷、脱发等）12 月多消退，仅各 1 例持续。
• 满意度：FACE-Q 外貌量表 12 月中位 Rasch 76（IQR 69–82），结果量表中位 73（63–87），两组较基线均显著提升（P=0.009 / P=0.004）；3 月达平台（消肿后）。结论：内镜面部年轻化并发症少、满意度高、技术安全。

Abstract 原文摘要

Background:Face lift techniques have greatly evolved in the last decades. Compared with more traditional methods, an endoscopic approach has the advantage of providing access to deeper structures while reducing extensive dissection of the face. The main objectives of this study are to assess the postoperative complications, aesthetic results, and patient satisfaction following 2 different face lift approaches.

Methods:We conducted a hybrid longitudinal observational study. According to patients’ preference, an upper two-thirds face lift (using a minimal endoscopic approach with blepharoplasty) or a full face lift was performed (including deep superficial musculoaponeurotic system dissection). Follow-up visits were scheduled up to 12 months postsurgery. At each visit, a standardized physical examination was conducted to assess sensory nerve function, and the House–Brackmann classification was used for motor nerve testing. The FACE-Q questionnaire was also administered to assess self-reported outcomes of satisfaction.

Results:A total of 35 patients underwent a facial rejuvenation procedure. Most sensory nerve disturbances resolved, except in 1 case, and all motor nerve disturbances had resolved by 12 months postsurgery. Other visible side effects had virtually disappeared at 12 months, with 1 persistent case of tightness and 1 of alopecia. Patient satisfaction regarding facial appearance and surgical results was high (overall median Rasch score of 76 and 73, respectively).

Conclusions:Endoscopic facial surgery can be used for facial rejuvenation with few surgical complications. Our results suggest that endoscopic facial surgery is safe and that patients are satisfied with this technique.

INTRODUCTION

Since the original face lift technique, developed by Skoog 1 in the 1970s, which involved lifting a single flap, face lift techniques have evolved to include steps to address deeper structures, providing more natural and lasting results. 2 Today, multiple variations of subcutaneous facial lifting with superficial musculoaponeurotic system (SMAS) plication, with or without excision, have emerged. 3 – 9 However, performing a preauricular incision involves more risk to the facial nerves and risk of skin necrosis. 10 Others have investigated “closed” methods, such as the subperiosteal rhytidectomy with posterior modification to reduce scarring but with limited aesthetic effects on the skin and neck region. 8 , 11 – 14 The endoscopic frontal face lift combined with an intraoral approach was also developed to better protect the temporal branch of the facial nerve in the zygomatic arch, but it is less common because this technique is difficult and carries risks. 15 , 16 Possible complications include significant temporal depression, weakening of the orbicularis oculi, and nerve trauma. 1 , 17 For the central third of the face, subciliary incisions or blepharoplasty combined with sub- or infraperiosteal dissection have also been described. 1 , 18 – 21

Adapting the work of other researchers, 1 , 15 , 22 – 24 we have begun using an endoscopic technique from the frontal region to the lower part of the face. Although the risk of trauma to the temporal and other facial branches remains, it can be mitigated with optimal knowledge of anatomy, proper incision planning, and meticulous dissection under direct endoscopic vision. The main benefit of this approach is the ability to reposition structures without extensive dissection, reducing the risk of complications. Therefore, to accommodate the needs of patients, 2 personalized options were offered to our patients: (1) endoscopic upper two-thirds face lift only or (2) full face lift with endoscopic assistance. This study aimed to assess the impact of these procedures on sensory and motor nerves, the aesthetic quality of the result, and the patients’ satisfaction with each procedure.

METHODS

Study Design and Setting

This is a hybrid observational longitudinal study conducted in a private surgical facility. The study is mainly prospective, with 3 retrospective cases. All measurements were done prospectively, and the retrospective cases were assessed using the same rigorous confidentiality rules. We used a nonprobabilistic convenience sample, and patients were recruited on a voluntary basis by their cosmetic surgeon. Data collection and measurements were standardized and were performed by the same person. All interventions were day procedures performed between January 2021 and December 2023. The study protocol was approved by the local ethics committee of the Centre intégré universitaire de santé et services sociaux de l’Estrie—Centre hospitalier universitaire de Sherbrooke.

Population

The target population comprised patients with the following inclusion criteria: presenting for facial rejuvenation, aged at least 35 years, with low anesthetic risk (American Society of Anesthesiology classification, ASA 1−2), and with realistic expectations. Those with preoperative facial or trigeminal nerve disturbance were excluded.

Study Interventions

All procedures were performed by the same surgeon and anesthetist. The choice of intervention was made with the patients, according to their needs. Patients were advised on the advantages and disadvantages of both techniques.

Endoscopic Upper Two-third Face lift Surgery

A 5-mm, 30-degree endoscope was used to make 5 incisions in the hairline. The dissection was carried out from the forehead through the traditional endoscopic forehead lift. 25 Dissection was carried down to the midface, as previously described. 22 , 24 Aiming toward the oral commissure, dissection was performed above the malar body to minimize the risk of injury to the temporal branch. Dissection was carried out under the same SMAS layer.

Then, dissection progressed to the midface, aiming toward the oral commissure and stopping at the nasal alae. A 1- to 2-cm wide tunnel was created to allow for the insertion of a deep suture and attachment to the superficial temporal fascia. Initially, fixation of the forehead was done with screws, as developed by Romo et al, 25 but this led to asymmetry in the earlier convalescence period. Therefore, we switched to cable suspension, as described by Massoud and Aboelatta, 26 to suspend the brow, which solved the problem. After deep SMAS dissection, plication was done with polydioxanone (PDS) 3-0 for the midface and fixed to the temporal fascia. A PDS 2-0 suture was used for the forehead cable suspension. 24 Traditional open upper and lower blepharoplasty was performed. Figure 1 illustrates the pattern of dissection and suture plication in an endoscopic lift.

Fig. 1. Pattern of dissection and suture plication to the temporal fascia in an endoscopic lift. White dotted line represents a 1–3 cm in length for the hairline incision; yellow dotted lines represent the extent of sub-SMAS dissection; black dots represent suture imbrication into the SMAS with PDS 3-0; white marks represent the temporal suspension of the PDS to the temporal fascia.内镜提升中颞部解剖与缝合折叠模式（全文"筋膜提升"核心图）。白虚线=发际 1–3 cm 切口；黄虚线=SMAS 下分离范围；黑点=PDS 3-0 将 SMAS 做叠瓦式折叠；白标记=PDS 悬吊至颞筋膜。

Full Face Lift Surgery With Endoscopy

This approach combined an endoscopic upper two-thirds face lift (as described earlier) with the traditional lower face lift. For the lower face, we started with the neck surgery. A submental incision was used. The subplatysmal work was executed as indicated for each patient: fat, the digastric muscle, and the submandibular gland were resected.

Central platysmal plication was done when indicated with PDS 3-0. 27 In the cheeks, pre- and retroauricular incisions were made. 4 Deep SMAS dissection and plication were performed with PDS 3-0. Fat grafting was added as indicated on a case-by-case basis.

Medication and Perioperative Care

Premedication consisted of administration of celecoxib and acetaminophen before surgery. Antinausea medication and cortisone were given perioperatively. Sedation consisted mainly of remifentanil and Versed. Nasal prongs were used to deliver oxygen. Incisions were infiltrated with 40 mL of 0.5% Xylocaine, with 1:400,000 epinephrine. Then, the remaining facial soft tissue was infiltrated with 120 mL of a tumescent solution composed of 0.05% Xylocaine and 1:1,000,000 epinephrine. Finally, 5 mL of 0.25% Marcaine with epinephrine was added at the end of the procedure. No antibiotics were given after the use of screws was discontinued.

Patients were discharged after 4–5 hours in the recovery room. They were seen the next day. Drains were removed at that time. No skin treatments, such as peeling or laser treatment, were performed.

Measures and Questionnaires

Before surgery, facial sensory testing was performed to establish a baseline, and patients completed the FACE-Q questionnaire, focusing only on the current appearance of their face. Subsequently, they attended follow-up visits at 1, 3, 6, and 12 months postsurgery to reassess these outcomes in addition to facial motor function, patient-reported side effects, and patient satisfaction with the results. Left and right sides were both evaluated in the same way.

Sensory Testing

Facial neurosensory testing was performed according to a method inspired by Meyer and Bagheri, 2 but the affected area was not delineated, as the entire face could be affected. Paresthesia was assessed at 3 levels (schematized in Fig. 2 ) for 7 different facial regions corresponding to specific facial nerves. In level A, directional discrimination and 2-point discrimination were assessed. A-1: Using a cotton-tipped applicator, smooth horizontal or vertical strokes were performed in each facial region. Good answers were considered normal (score 1), and bad answers were considered abnormal (and the test proceeded to level B for this region). A-2: For each facial region, the 2-point discrimination at distances of 0, 5, 10, 15, and 20 mm was assessed using a caliper, and the highest normal response was noted. If the difference between the baseline and follow-up assessments was higher than or equal to 10 mm, the result was considered abnormal for that region, and the test continued to level B. Otherwise, the region was considered normal (score 1), and no more testing was done. In level B, a wooden stick was lightly applied to each facial region, and the patient was asked to confirm when they felt the contact, which was considered a normal response (score 2 = mild hypoesthesia). If the patient did not feel the first contact, a second contact was applied with more pressure, enough to observe an indentation in the skin. If the patient perceived the increased pressure or did not feel anything, this was considered an abnormal result, and the test proceeded to level C. In the final level, a 27G needle was used to repeat the same test as in level B. First, the patient was asked to confirm whether they felt the light touch of the needle. If they did, the test was over (score 3 = moderate hypoesthesia). If they did not feel the contact, the touch was repeated with a little more pressure using the needle, enough to produce an indentation in the skin without piercing it. Whether they felt the contact of the needle (score 4 = severe hypoesthesia) or not (score 5 = anesthesia), the test was then over. It is worth adding that throughout this testing, a few fake contacts were added to mislead the patients and ensure that they provided real responses to the test.

Fig. 2. Schematic representation of sensory testing. At level A, the test is performed with a cotton-tip applicator (A-1), and the second step uses a caliper (A-2). For levels B and C, the tests are performed with a wooden stick (B) and a 27G needle (C). The 7 facial regions are identified by gray zones. AT, auriculotemporal; M, mental; IO, infraorbital; SO, supraorbital; ST, supratrochlear; ZF, zygomaticofacial; ZT, zygomaticotemporal). Some items in this figure were designed using resources from Flaticon.com.感觉神经测试方案示意（改良 Meyer & Bagheri 法）。A 级用棉签定向辨别（A-1）与卡尺两点辨别（A-2，0/5/10/15/20 mm）；异常者进入 B/C 级进一步评估。覆盖 7 个面部区域对应各面神经分支。

Motor Nerve Testing

For the motor nerve testing, the House–Brackmann classification was used. 6 The surgeon assessed 3 regions separately: the forehead (temporal branch of the facial nerves, grades I–IV), the eyes (zygomatic branch of the facial nerves, grades I–V), and the mouth (buccal branch of the facial nerves, grades I–VI). Grade I designated normal motor functions, whereas the other grades were considered abnormal (the highest grades representing paralysis).

Patient-reported Side Effects

Patient-reported side effects were recorded by the surgeon on the patient evaluation form during the physical examination. Facial deformity for 7 facial regions (forehead, temple, sides, cheeks, upper lip, chin, and jaw), tightness in 3 regions (forehead, temple, and sides), and alopecia (according to Ludwig grades I–III) were assessed. 28 , 29 On the other hand, pain in each region (forehead, temple, sides, cheeks, upper lip, chin, and jaw), headache, and difficulty chewing were assessed by questioning the patients.

Patient Satisfaction

The Appearance and Results scales of the FACE-Q questionnaire were used to collect self-reported satisfaction with rejuvenation outcomes. 7

Statistical Analysis

Categorical variables were presented using frequencies and proportions. When indicated, the unit for the frequencies and proportions was the facial sides (left and right), that is, double the number of patients. Continuous variables were described using medians and interquartile ranges (IQRs), as visual inspection of the histograms revealed an asymmetrical distribution. Complete-case analyses were performed for each variable of interest; no imputation was done. Mann-Whitney U tests (for continuous variables) and χ 2 or Fisher exact tests (for categorical variables) were used to compare the 2 types of interventions. The longitudinal satisfaction scores were presented graphically using locally estimated scatterplot smoothing curves along their 95% confidence intervals, using a default parameter (span = 0.75 and degree = 2). Participants’ own appreciation of their facial appearance was compared within each group separately between baseline and the 12-month follow-up using a (paired) Wilcoxon signed-rank test. All analyses were done using R, version 4.5.1.

RESULTS

Surgery Safety

A total of 35 patients underwent one or the other surgical intervention, with about half in each group. All patients were women, with a median age of 58 years, ranging from 41 to 76 years. A trend was observed as younger patients tended to choose an endoscopic-only intervention. Unsurprisingly, those who chose the full face lift had a longer surgery time (Table 1 ).

Table 1. Characteristics of Participants, Including Postsurgical Testing and Patient-reported Side Effects

The supratrochlear nerves were most frequently affected, whereas the mental nerves were the least affected. Complete resolution of sensory nerve disturbance was observed at 12 months postsurgery, except for the supratrochlear nerves in 1 patient who experienced lasting mild hypoesthesia (Table 2 ).

Table 2. Prevalence of Sensory Nerve Disturbance (Merged Results for Both Right and Left Sides)

Motor nerves were affected in 24 (82.8%) patients (Table 1 ), and the temporal branch was the most frequently impacted (Table 3 ). Complete resolution of motor deficits was observed at 12 months. Other side effects were observed after surgery, and most of them were resolved after 12 months (Table 4 ). We did not find any statistical differences between the approaches in terms of postsurgical complications (side effects, sensory nerve disturbance, and motor nerve disturbance) except for the buccal branch of the facial nerve, which was more frequently affected (50% of cases) at 1 month in patients undergoing the full face lift ( P = 0.0003).

Table 3. Prevalence of Motor Nerve Disturbance (Merged Results for Both Right and Left Sides)

Table 4. Prevalence of Other Side Effects of Surgery

Patient Satisfaction

The FACE-Q questionnaire responses were used to measure rejuvenation outcomes. Results were available for all patients at least once, and data were available in 57%–68% of cases over time. The patients’ level of satisfaction regarding their facial appearance increased significantly from baseline to the 12-month evaluation in each group separately, reaching a global median (IQR) Rasch score of satisfaction of 76 (69–82) at term (Fig. 3 ). For those who chose the upper two-thirds face lift (line A) and those who chose a full face lift (line B), the increase in satisfaction at 12 months was statistically significant (respectively, P = 0.009 and P = 0.004 compared with baseline). The score given by patients to indicate their level of appreciation of the results of the surgery improved slightly over time, reporting a median (IQR) satisfaction of 73 (63–87) at 12 months postsurgery (Fig. 4 ). There was no significant difference between the 2 types of interventions at any time during the 12 months of follow-up for both scores.

Fig. 3. Patients’ appreciation of their facial appearance before and after surgery (Rasch score). “A” corresponds to patients who chose the upper two-thirds lift (endoscopic-only intervention), whereas “B” corresponds to patients who chose a full face lift.FACE-Q 外貌量表（Appearance）术前 vs 术后 Rasch 评分。两组（A=上 2/3 内镜，B=全脸）12 月均显著提升，全局中位 76（IQR 69–82）；3 月后趋平台（消肿后正性印象确立）。

Fig. 4. Patients’ appreciation of the surgery results before and after surgery (Rasch score). “A” corresponds to patients who chose the upper two-thirds face lift (endoscopic-only intervention), whereas “B” corresponds to patients who chose a full face lift.FACE-Q 结果量表（Results）术前 vs 术后 Rasch 评分。12 月中位 73（IQR 63–87），随时间略升；同样在 3 月达平台。

DISCUSSION

In our study, we used 2 approaches, depending on patients’ preferences and needs. The first technique refers to an endoscopic upper two-thirds face lift equivalent to those reported by Kao and Duscher 24 (ponytail lift type 1), Firat 30 (group 1), and Gennai et al 31 (Mivel 2). The second technique consisted of a full face lift combining the first technique (endoscopic surgery) with an open deep SMAS dissection, as previously described by Casagrande et al, 22 Kao and Duscher 24 (posterior talofibular ligament-II), and Firat 30 (group 4). The decision to choose 1 technique over the other was made by the patients after counseling by the surgeon and based on the area they wished to rejuvenate.

The endoscopic upper two-thirds face lift technique was adopted more frequently by younger women who did not have an excess of skin. In patients who had an excess of skin, it was redistributed to the scalp area. There is no consideration for the length of the forehead. Lifting involves a lateral upper rotation of the tail of the eyebrow without increasing the midline forehead length. This technique does not require extensive open dissection while still treating the whole face. Indeed, it provides maximum results with minimal scarring while allowing middle and lower face improvement to different degrees. This technique can still be offered to patients who do not want to treat the lower part of the face. For patients with an excess of skin or with significant aging in the lower third of the face and neck, this technique can be combined with an open dissection. 22 The best natural results are usually observed when encompassing the whole face (Fig. 5 ), although patients who do not want the full-face procedure can be satisfied with an endoscopic upper two-thirds face lift (Fig. 6 ). We observed that both techniques seem equivalent in terms of appreciation by the patients (Figs. 3 , 4 ) and in terms of safety (Table 1 ).

Fig. 5. A 68-year-old woman who had a full face lift with endoscopy. A, Preoperative. B, One year postoperatively.68 岁女性，行全脸提升（内镜上 2/3 + 开放下脸深部 SMAS）。A 术前；B 术后 1 年。可见中下面部提升与下颌轮廓改善。

Fig. 6. A 71-year-old woman who had an endoscopic upper two-thirds face lift. A, Preoperative. B, Six months postoperatively.71 岁女性，行内镜上 2/3 面部提升。A 术前；B 术后 6 月。眉尾外上旋转、中面部轻度提升，瘢痕隐蔽于发际。The use of blepharoplasty in all cases also contributed to the overall appearance. The lower blepharoplasty improved wrinkles with an upward skin shift, whereas the upper blepharoplasty minimized the impact of forehead relapse on the eyelid. Multiple minimally invasive techniques have been described, each with its own limitations. 3 , 5 , 8 , 9 , 11 – 13 Our approach is less invasive than a subperiosteal face lift, as previously described. 14 , 32 We reported a high rate of facial nerve disturbance (up to 82.8% at 1 mo postsurgery), whereas others have reported 0.8%−5% of transient neuropraxia. 15 , 16 , 24 The difference in these results might be due to the timing of the measurement, the assessment technique, and the criteria for defining impairment. However, in all reports, including ours, most cases resolved spontaneously. In our study, we used the House–Brackmann classification to report motor nerve anomalies, where any disturbance was considered abnormal. Following surgery, only a few retouches were required: 1 case of lipoinjection, 1 neck revision, 4 eyelids, and 11 cases benefited from injections with hyaluronic acid and neuromodulators.

This study has limited external validity. Although the study population, the endoscopic tools, and the techniques used are certainly transferable to other clinical settings, the procedures were performed by a single surgeon. Thus, we could not measure inter-rater reliability, and this could have biased the study. The hybrid design may increase the variability of the data. Other limitations include the small number of patients and the novelty of the procedure in our clinical setting. Moreover, although we used a validated questionnaire to assess patient satisfaction, the appearance of self-appreciation remains subjective and difficult to compare over time. Some data were missing due to missed follow-up visits. This study also has strengths. It was done mostly prospectively with a homogenous population and a high retention rate of up to 1 year postsurgery. Rigorous objective methods were used to assess sensory and motor nerve disturbances and patient satisfaction.

Our results suggest that these 2 endoscopic face lift approaches may be satisfactory and safe. We did not aim to assess the superiority of one technique over the other, as this would have required a larger sample size. We reported detailed postsurgical nerve disturbances and side effects, which mostly resolved within a year. Patients’ satisfaction with facial appearance and surgery results was high. Although we observed a plateau at 3 months, it reflects patients’ positive impression after the swelling subsided. Therefore, we are confident that these endoscopic techniques for facial rejuvenation are safe to be performed in both younger and older women. The safety and level of patient satisfaction following these endoscopic approaches should encourage clinical settings that already use endoscopy for frontal surgery to extend this technique to face lift surgery.

DISCLOSURE

The authors have no financial interest to declare in relation to the content of this article.

PATIENT CONSENT

Patients provided written consent for the use of their images.

ACKNOWLEDGMENTS

The authors thank the Memorial Sloan Kettering Cancer Center (New York, NY), which holds the copyright to the FACE-Q questionnaire and its translations. They also wish to thank Jeanne Pratte, Amélie Tétu, and Catherine Allard from the Unité de Recherche Clinique et Épidémiologique at the Centre de recherche du CHUS for their support with article revision and statistical analyses. The authors thank Bonita Van Doorn for the linguistic revision of the article. The authors are very grateful to Sonia Cheng Oviedo for her advice and support, and to the members of the Centre de chirurgie esthétique Sherbrooke for their help during the clinical visits.

临床落地解读

"筋膜提升"的核心落点： 本文两种术式的提升主体都是 深部 SMAS 解剖 + PDS 折叠/悬吊 ——这正是"筋膜提升"的本质。 内镜上 2/3 入路在 SMAS 同层下分离，用 PDS 3-0 将 SMAS 做叠瓦式（imbrication）折叠并悬吊固定至 颞筋膜 ； 全脸入路则在下脸经耳前/耳后切口行开放深部 SMAS 解剖与折叠，并对颈阔肌中央折叠。换言之，无论内镜还是开放， "挂住并重新定位筋膜"才是抗衰持久的关键，而非单纯拉皮。

内镜微创化的价值与边界： 经发际 5 mm 小切口、在直视下于 SMAS 层内精细分离，可 减少广泛剥离 、 降低皮瓣坏死与感觉障碍范围，对皮肤不松弛的较年轻患者尤适合（文中内镜组更年轻、手术时长仅 2.9 h vs 全脸 4.7 h）。 但对皮肤冗余、下面部-颈部老化明显者，作者明确指出需联合开放下脸 SMAS/颈阔肌处理才能获得自然结果——内镜不是万能替代。

神经安全信号须冷静看待： 本研究术后 1 月运动神经障碍率 82.8%，远高于文献报道的 0.8%–5% 暂时性神经失用。 作者将其归因于 评估时点更早、测试方法更细、判定标准更严 （House–Brackmann 分区分级 + 改良感觉测试）， 且所有病例 12 月内恢复。这提示：内镜 SMAS 操作虽微创，但颞支、颊支仍处风险区， 必须依赖扎实解剖、切口规划与直视下锐性/钝性分离；报道的"高发生率"更多反映检测灵敏度，而非真实损伤率。

证据等级与适用边界： 样本仅 35 例、 全部为女性 、单术者、含 3 例回顾性、满意度为自报量表（主观）， 且作者明确"未比较两种术式优劣"（需更大样本）。因此本文价值在于 提供一套可复用的内镜 SMAS 提升安全-满意度基线数据 ， 而非证明某术式更优。临床落地时，应将其视为技术可行性与安全监测框架，结合个体解剖与老化量表选择内镜或联合开放路径。

与既往日更的呼应： 本系列已涵盖 HA 填充血管安全、线雕、能量设备（HIFU/射频）等微创方向； 本文补上 手术层面"筋膜提升"（SMAS）的循证拼图 ——它代表了微创注射与能量设备之上、针对皮肤松弛的更深层干预， 三者在"由浅入深"的抗衰光谱中各有定位。