https://orcid.org/0000-0003-0952-8090
Figures
Abstract
Objective
The Da Vinci Robot is the most advanced micro-control system in endoscopic surgical instruments and has gained a lot of valuable experience today. However, the technical feasibility and oncological safety of the robot over open surgery are still uncertain. This work is to systematically evaluate the efficacy of the unilateral axillary approach for robotic surgery compared to open surgery for differentiated thyroid carcinoma.
Methods
PubMed, Embase, Cochrane Library, and Web of Science databases were utilized to search for relevant literatures of robotic thyroid surgery using unilateral axillary approach compared to open thyroid surgery, and a meta-analysis was performed using RevMan software version 5.3. Statistical analysis was performed through Mantle-Haenszel and inverse variance methods.
Results
Twelve studies with a total of 2660 patients were included in the meta-analysis. The results showed that compared with the open group, the robotic group had a longer total thyroidectomy time, shorter hospital stay, less intraoperative bleeding, more postoperative drainage, fewer retrieved central lymph nodes, and higher cosmetic satisfaction (all P < 0.05). In contrast, temporary and permanent laryngeal recurrent nerve injury, temporary and permanent hypoparathyroidism or hypocalcemia, brachial plexus nerve injury, number of retrieved central lymph nodes, number of retrieved lymph nodes in the lateral cervical region, number of lymph node metastases in the lateral cervical region, hematoma, seroma, lymphatic leak, stimulated thyroglobulin (sTg) and unstimulated thyroglobulin (uTg), and the number and recurrence rate of patients with sTg <1ng/ml were not statistically different between the two groups (P > 0.05).
Citation: Zhang X, Yu J, Zhu J, Wei H, Meng N, Hu M, et al. (2024) A meta-analysis of unilateral axillary approach for robotic surgery compared with open surgery for differentiated thyroid carcinoma. PLoS ONE 19(4): e0298153. https://doi.org/10.1371/journal.pone.0298153
Editor: Laura Hannah Kelly, Public Library of Science, UNITED KINGDOM
Received: September 12, 2023; Accepted: January 19, 2024; Published: April 11, 2024
Copyright: © 2024 Zhang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript. The included studies are available from the PubMed, Embase, Cochrane Library, and Web of Science databases.
Funding: NO.
Competing interests: NO.
1 Introduction
Thyroid cancer is one of the most common endocrine-related cancers, and papillary and follicular carcinomas are collectively referred to as differentiated thyroid carcinomas (DTC). Compared to open thyroidectomy (OT), endoscopic thyroidectomy (ET) avoids neck scarring and has excellent cosmetic results, making it popular among patients. However, the scope of ET is affected by the limitations of rigid instruments. Robotic thyroidectomy (RT), which has the advantage of both "flexible" robotic arms and aesthetic appearance of the neck, is expected to overcome the shortcomings of ET and be gradually carried out all around the world. Currently, several approaches to RT have been used in clinical practice, including the unilateral axillary, bilateral axillary breast, transoral, and retroauricular approaches [1]. Because of the small number of cases in each study, the efficacy cannot be objectively evaluated, and most of the existing meta-analyses have covered different approaches, resulting in high heterogeneity and affecting the credibility of conclusions [2–4]. Therefore, this study intends to conduct a meta-analysis of RT versus OT in a single approach (unilateral axillary approach), and systematically evaluate their surgical efficacy as below.
2 Materials and methods
2.1 Literature search
PubMed, Embase, Cochrane Library, and Web of Science databases were used for clinical literature research on the comparison of RT and OT. Keywords used were "robotic", "open", "conventional", "thyroidectomy", and "neoplasms". The retrieval time is from the establishment of the database to April 2023. The search strategy and PICOS strategy are shown in S1 Table.
2.2 Inclusion criteria
(1) randomized, non-randomized, case-control, cohort, and cross-sectional studies of RT and OT; (2) the surgical approach of RT is limited to the unilateral axillary approach; (3) postoperative pathology of patients confirmed to be DTC; and (4) at least one surgery-related outcome indicator can be provided.
2.3 Exclusion criteria
(1) history of neck surgery or radioactive iodine therapy; (2) repeated publication of literature, case reports, conference papers, expert comments, reviews, etc.; and (3) lack of original data and fruitless contact with authors.
2.4 Data extraction and literature quality evaluation
Two researchers independently screened the literature and extracted the relevant information. In case of disagreement, a third researcher was solicited to make the final decision. The extracted data included the following terms: basic information of the included studies (including the author, publication year, country, number of surgeries, sex, age, surgery procedure, follow-up time, study types, and quality scores) and 21 outcomes related to surgery (including temporary and permanent recurrent laryngeal nerve injury, temporary and permanent hypoparathyroidism or hypocalcemia, brachial plexus injury, number of retrieved central lymph nodes, number of central lymph node metastases, number of dissected lymph nodes in the lateral neck region, number of metastatic lymph nodes in the lateral neck region, time of total thyroidectomy, length of hospital stay, intraoperative blood loss, postoperative drainage, hematoma, seroma, lymphatic leakage, level of stimulated thyroglobulin (sTg) and unstimulated thyroglobulin (uTg), number of people with sTg <1ng/ml, recurrence rate, and cosmetic satisfaction rate). The criteria for temporary/permanent recurrent laryngeal nerve injury, temporary/permanent hypoparathyroidism, and hypocalcemia were determined according to the duration of each study. The rate of recurrent laryngeal nerve injury was calculated as follows: number of recurrent laryngeal nerve injuries/total number of recurrent laryngeal nerve exposure ×100%. The sTg levels were compared only in patients treated with radioactive iodine after total thyroidectomy. Cochrane criteria, Methodological Index for Non-Randomized Studies(MINORS), Newcastle-Ottawa scale(NOS), and Healthcare Research and Quality(AHRQ) were used to evaluate the quality of literature for randomized controlled studies, non-randomized controlled studies, cohort studies, and cross-sectional studies, respectively.
2.5 Statistical methods
RevMan software (version 5.3) was used for the meta-analysis. Statistical analysis was performed through Mantle-Haenszel and inverse variance methods. The risk ratio (RR) was used as the effect indicator for dichotomous variables, and the weighted mean difference (MD) or standardized mean difference was used for continuous variables. Random-effects models were used, with estimates and 95% confidence intervals (CIs) provided for all indicators. The test level for the meta-analysis was α = 0.05, and P < 0.05 was set as statistically significant.
3 Results
3.1 Literature retrieval results
After literature searching, 766 articles were initially retrieved, one article was added by tracing the original text through the references, 337 duplicate articles were removed, 396 were screened out by reading the titles and abstracts, 22 articles were excluded in strict accordance with the inclusion and exclusion criteria, and 12 studies were finally included in this meta-analysis. A flowchart of the search process is shown in Fig 1.
3.2 Basic characteristics and quality assessment of the included studies
A total of 12 publications were included in this meta-analysis [5–16], including three non-randomized controlled studies, which had a MINORS score of 17,21,20 separately [5, 7, 9], eight retrospective case-cohort studies, which had a NOS score of 7,8,8,8,9,9,8,9 separately [6, 8, 11–16], and one cross-sectional study, which had a AHRQ score of 18 [10]. In addition, 10 of the articles were from Korea [5–7, 9–14, 16] and 2 from the United States [8, 15]. There were a total of 2660 patients (1208 in the RT group and 1452 in the OT group), among which the proportion of female patients was 76% in the RT group and 72% in the OT group, while the proportion of total thyroidectomy was 65% in the RT group and 77% in the OT group (Tables 1 and 2). Whereas 22 articles excluded after reviewing the full text is shown in Table 3 [17–38].
3.3 Meta-analysis results
3.3.1 Postoperative complications.
Transient recurrent laryngeal nerve injury. A total of eight studies [5–8, 10, 12, 13, 15] were used to study temporary recurrent laryngeal nerve injury with low heterogeneity (I2 = 0%), wherein the injury proportion of 1.57% (13/826) in the RT group was slightly lower than that of 2.03% (24/1184) in the OT group, with no statistically significant difference (RR = 0.81; 95% CI:0.41–1.61; P = 0.55) (Fig 2A).
RT, robotic thyroidectomy; OT, open thyroidectomy; M-H, Mantel–Haenszel; CI, confidence interval.
Permanent recurrent laryngeal nerve injury. A total of seven studies [6, 7, 10, 12–15] were included with low heterogeneity (I2 = 0%), and the difference was not statistically significant (RR = 1.22; 95% CI:0.40–3.74; P = 0.73). The proportion of injury in the RT group was 0.38% (3/785), which was slightly lower than that in the OT group (0.56%; 7/1240) (Fig 2B).
Brachial plexus injury. A total of three studies [7, 12, 14] were included, and only one study reported one case of brachial plexus injury in the RT group. The difference was not statistically significant (RR = 7.36; 95% CI:0.31–176.99; P = 0.22) (Fig 2C).
Transient hypoparathyroidism or hypocalcemia. Nine studies [5–8, 10, 12–15] were included with low heterogeneity (I2 = 0%). There was no statistically significant difference between the symptom incidences of 27.3% (140/513) in the RT group and 31.0% (235/757) in the OT group (RR = 0.90; 95% CI:0.76–1.07; P = 0.23) (Fig 3A).
Forest plots displaying incidences of (A) transient hypocalcemia or hypoparathyroidism; (B) permanent hypocalcemia or hypoparathyroidism; (C) hematoma; (D) seroma; (E) lymphatic leakage RT, robotic thyroidectomy; OT, open thyroidectomy; M-H, Mantel–Haenszel; CI, confidence interval.
Permanent hypoparathyroidism or hypocalcemia. Six studies [6, 7, 10, 12–14] were included with low heterogeneity (I2 = 0%). There was no significant difference (RR = 1.32; 95% CI:0.30–5.78; P = 0.71) between the RT (0.73%; 3/413) and OT (0.53%; 3/570) groups (Fig 3B).
Hematoma. Seven studies [5–7, 10, 12–14] were included with low heterogeneity (I2 = 0%). There was no significant difference (WMD = 0.75; 95% CI:0.31–1.83; P = 0.53) between the 1.5% (7/454) of hematomas in the RT group and 2.1% (13/613) in the OT group (Fig 3C).
Seroma. Seven studies [5, 6, 10, 12–15] were included with low heterogeneity (I2 = 1%). There was no significant difference (WMD = 1.25; 95% CI:0.79–1.96; P = 0.34) in the incidence of seroma in the RT (9.1%; 39/427) and OT (7.2%; 47/656) groups (Fig 3D).
Lymphatic leakage. Five studies [6, 7, 12, 14, 15] were included with low heterogeneity (I2 = 0%). There was no significant difference (WMD = 1.21; 95% CI:0.61–2.41; P = 0.58) in lymphatic leakage between the RT (5.9%; 13/219) and OT (5.4%; 23/427) groups (Fig 3E).
3.3.2 Oncological outcomes.
Number of retrieved central lymph nodes. Six studies [5, 6, 9, 12, 13, 16] were included and divided into unilateral [5, 9, 12, 16] and bilateral [6, 12, 14, 16] subgroups according to the scope of dissection. Heterogeneity was low in both subgroups (I2 = 0%), but there was a significant difference in the number of retrieved central lymph nodes between the two groups. In the unilateral group, there was no significant difference between the RT and OT groups (WMD = -0.36; 95% CI: -0.86–0.14; P = 0.16). However, the number of retrieved central lymph nodes in the bilateral RT group was significantly lower than that in the OT group (WMD = -1.77; 95% CI: -2.79 –-0.76; P = 0.0006) (Fig 4A).
Forest plots displaying number of (A) retrieved central lymph nodes; (B) central lymph node metastases; (C) retrieved lymph nodes in the lateral cervical region; (D) metastatic lymph nodes retrieved from the lateral component RT, robotic thyroidectomy; OT, open thyroidectomy; SD, standard deviation; CI, confidence interval; IV, inverse variance.
Number of central lymph node metastases. A total of six studies [5, 11–14, 16] were included with low heterogeneity (I2 = 25%). There was no significant difference between the two groups (WMD = 0.21; 95% CI: -0.02–0.43; P = 0.07) (Fig 4B).
Number of retrieved lymph nodes in the lateral cervical region. Four studies [6, 7, 12, 14] were included with low heterogeneity (I2 = 0%). There was no statistically significant difference between the RT and OT groups (WMD = -0.65; 95% CI: -2.84–1.54; P = 0.56) (Fig 4C).
Number of metastatic lymph nodes in the lateral cervical region. Two studies [12, 14] were included with low heterogeneity (I2 = 0%). There was no statistically significant difference between the RT and OT groups (WMD = -0.49; 95% CI: -1.70–0.71; P = 0.42) (Fig 4D).
3.3.3 Operation-related events.
Operation time of total thyroidectomy. A total of eight studies [5–9, 12–14] were included for comparison of operation time. According to the scope of lymph node dissection, patients were divided into the central lymph node dissection [5, 8, 9, 13] and functional neck dissection [6, 7, 12, 14] subgroups. Heterogeneity was low in both groups at 30% and 0%, respectively. In the two subgroups, the operation time of the RT group was significantly longer than that of the OT group, and the differences in the central lymph node dissection subgroup (WMD = 29.04; 95% CI:20.78–37.30; P < 0.00001) and functional neck dissection subgroup (WMD = 58.30; 95% CI:48.99–67.60; P < 0.00001) were both statistically significant (Fig 5A).
Forest plots displaying (A) operative time of total thyroidectomy; (B) length of hospitalization; (C) drain amount; (D) amount of intraoperative bleeding RT, robotic thyroidectomy; OT, open thyroidectomy; SD, standard deviation; CI, confidence interval; IV, inverse variance; TT, total thyroidectomy; CCND, central component lymph node dissection; MRND, modified radical lymph node dissection.
Length of hospitalization. Nine studies [5–9, 11, 12, 14, 15] were included with high heterogeneity between the groups (I2 = 80%). The length of hospitalization in the RT group was significantly shorter than that in the OT group (WMD = -0.58; 95% CI: -0.93 –-0.23; P = 0.001) (Fig 5B).
Volume of drainage. Two studies [12, 13] were included with low heterogeneity (I2 = 0%). The RT group resulted in significantly more volume of drainage than the OT group (WMD = 84.59; 95% CI:69.64–99.55; P < 0.00001) (Fig 5C).
Amount of intraoperative bleeding. Three studies [5, 8, 15] were included with low heterogeneity (I2 = 0%). The results showed that the amount of blood loss in RT group was significantly less than that in OT group (WMD = -1.58; 95% CI: -2.96 –-0.20; P = 0.03) (Fig 5D).
3.3.4 Postoperative events.
sTg (ng/ml) level. Seven studies [7, 8, 12–16] were included with high heterogeneity (I2 = 58%). There was no significant difference (WMD = -0.02; 95% CI: -0.18–0.14; P = 0.80) in the sTg levels between the RT and OT groups (Fig 6A).
Forest plots displaying (A) sTg; (B) number of sTg<1ng/ml; (C) uTg; (D) recurrence rate; (E) cosmetic satisfaction RT, robotic thyroidectomy; OT, open thyroidectomy; SD, standard deviation; CI, confidence interval; IV, Inverse Variance.
sTg < 1ng/ml. Only two studies [7, 16] had low heterogeneity (I2 = 0%). There was no significant difference (WMD = 0.98; 95% CI:0.94–1.03; P = 0.41) between the RT (93.4%; 184/197) and OT groups (95.2%; 260/273) (Fig 6B).
uTg (ng/ml) level. A total of four studies [6, 11, 14, 16] were included with low heterogeneity (I2 = 0%). There was no significant difference (WMD = -0.04; 95% CI: -0.33–0.25; P = 0.80) between the RT and OT groups (Fig 6C).
Recurrence. Seven studies [8, 11–16] were included. The recurrence rate was divided into two subgroups according to the follow-up periods of 2 and 5 years. The heterogeneity between the combined groups was low (I2 = 0%), and there was no significant difference (WMD = 1.06; 95% CI: 0.49–2.31; P = 0.88) in the recurrence rate of the RT (1.5%; 14/960) and OT (1.5%; 17/1122) groups (Fig 6D).
Cosmetic satisfaction. Two studies [5, 7] were included with high heterogeneity (I2 = 61%). The cosmetic satisfaction of the RT group was 90% (93/103), which was significantly (WMD = 1.38; 95% CI:1.06–1.80; P = 0.02) higher than that of the OT group (66%; 72/109) (Fig 6E).
4 Discussion
Generally speaking, DTC tends to occur in women between 40 and 50 years of age; however, the current trend is becoming younger [15]. Since the Da Vinci system was first used in thyroid surgery, different approaches have been proposed, which can be divided into two main approaches according to the operating direction: upward, bilateral axillo-breast approach, and unilateral axillary approach, as well as downward transoral and retroauricular approach [39]. Among these approaches, the transoral and retroauricular approaches are more difficult to operate because of the opposite surgical field due to the downward direction of the operation. In particular, the retroauricular approach has a small operating space and is difficult to perform due to the contralateral thyroid gland, resulting in contralateral incisions for total resection. In addition, the bilateral axillo-breast approach is difficult in patients with thoracic deformities and prominent clavicle protrusions. This method requires the injection of carbon dioxide to maintain the operating space, which may lead to hypercapnia, subcutaneous emphysema, and even gas embolism. At the same time, a large range of subcutaneous tissue separation is required because of the four incisions, which have a great impact on the sensation of the breast and chest wall. Compared to the bilateral axillo-breast approach, the unilateral axillary approach has fewer incisions and a smaller flap-free area [21].
Owing to these advantages, an increasing number of patients choose the unilateral axillary approach. Based on the experience of 5000 cases of unilateral axillary RT, Kim et al. [40] believed that this surgical procedure has the potential to reduce perioperative complications and achieve better surgical and oncological outcomes. To date, several early systematic reviews and meta-analyses of RT versus OT [2–4, 41–48] have been conducted, but the included studies were mixed with a variety of approaches, resulting in significant differences in the baseline characteristics of patients, high heterogeneity of results, and repeated inclusion of the same patient population, which affected the credibility of the results, which are shown in S2 Table. In this study, 12 included studies were strictly screened, and propensity score matching was used for four of them [11, 13, 14, 16]. It has been reported that propensity score matching can reduce selection bias, improve accuracy, and allow comparisons between different surgical procedures in nonrandomized retrospective studies [49, 50].
Compared with OT, the safety of unilateral axillary approach RT was the most important. The results of this meta-analysis showed that intraoperative blood loss in the RT group was significantly reduced compared to that in the OT group (P < 0.05). This is because the robotic system is equipped with 3D high-definition lens equipment, which can enlarge the surgical field of vision by 10–15 times, enabling the surgeon to identify fine vessels more clearly, enabling more precise hemostasis, and reducing the occurrence of bleeding.
However, the RT group had more postoperative drainage, and the difference was statistically significant compared to the OT group. Tae [13] believes that the range of RT-separated flaps is approximately three times that of OT, including the lateral neck and anterior chest wall. As the exudate would increase with the increase in surgical scope, this may be the reason for the significant increase in drainage after RT.
In addition, the total resection times in the two groups were significantly different. The patients were divided into two subgroups according to the scope of lymph node dissection because of the high heterogeneity, showing low heterogeneity in both groups. Regardless of the dissection method, the operation time in the RT group was significantly longer than that in the OT group. This may be related to the RT operation process, which includes three steps: creating the operation space, instrument docking, and the surgical operation. During the operation, the dense tissue in the neck region and lack of natural space made it necessary to conduct free dissection of the skin flap first, and the complex instrument docking adjustments and difficulty in the treatment of the contralateral thyroid lobe led to longer RT time. Lee et al. [7] found that the operation time of RT was similar to that of OT if the cavity construction and docking times were excluded, and most surgeons generally believed that the total operation time might be gradually shortened with the proficiency of surgical techniques [51, 52].
Another difference between the RT and OT groups was the length of the hospital stay. Some scholars believe that the discharge standards are that patients can walk freely after surgery, take drugs orally, tolerate pain, and have no surgically related complications [15]. However, Lee et al. [5] suggested that the drainage tube should be safely pulled out after surgery before discharge. This study showed that hospital stay in the RT group was shorter than that in the OT group, which may indicate that RT is safe and reliable. The high heterogeneity may be related to the different definitions of discharge criteria by surgeons; thus, the results should be treated with caution.
Except for the obvious differences mentioned above, there were no significant differences in most complications between the RT and OT groups (P > 0.05), including temporary and permanent recurrent laryngeal nerve injury, temporary and permanent hypoparathyroidism or hypocalcemia, brachial plexus nerve injury, hematoma, seroma, or lymphatic leak. Unlike manually controlled lenses in ET, the robotic system has a more stable field of vision, alleviates the surgeon’s sense of visual wobble and fatigue, and can better understand depth and distance. When dealing with the contralateral thyroid lobe, the camera could be rotated to ensure the surgical field of vision. For instance, if a deeper thyroid gland or forward protrusion of the trachea obstructs the operation, tilting the operating table contralateral upward by 10° to 15° can better expose the contralateral tracheoesophageal groove. Combined with the highly flexible manipulator, the surgeon is able to finely dissect the posterior thyroid peritoneum during the surgical operation, which facilitates the identification and protection of the recurrent laryngeal nerve and parathyroid glands, and thus may achieve the same results as open surgery [53].
In addition to safety, complications, and other variables, the completeness of total thyroidectomy has been of great concern and can be estimated postoperatively by ultrasound, uTg, sTg levels and whole-body scan iodine uptake levels. Since Tg can only be produced by normal thyroid or differentiated thyroid cancer cells, the clinical detection of Tg is simple, sensitive, and less invasive, and is considered a reliable indicator to measure the amount of residual tissue after total thyroidectomy. It has been proven that sTg less than 1 ng/ml after radioiodine therapy is a good marker for complete thyroidectomy [54]. Our study showed that the RT and OT groups had similar results with no statistical difference.
In addition, the number of cervical lymph node dissections is the main index for evaluating the thoroughness of surgery for DTC. Some studies have been conducted, but the high heterogeneity in most of them affects the reliability of the results [2, 4, 42]. In this study, heterogeneity was significantly reduced by dividing central lymph node dissection into unilateral and bilateral subgroups. The results showed no significant difference in the number of lymph nodes dissected from the unilateral central region between the two groups. However, when dissecting both sides, fewer lymph nodes were obtained in the RT group than in the OT group, and this difference was statistically significant. On one hand, obstruction of the trachea may be the reason. In this case, the field of vision and manipulation of instruments were limited, especially in the dissection of the posterior lymph nodes of the contralateral recurrent laryngeal nerve. However, some early published studies included in this study may be in the early exploratory stage of this operation, and the patients enrolled in RT belong to the early stage of the disease, resulting in a small number of dissected lymph nodes. In the future, lymph nodes may be removed more thoroughly as the operator becomes more proficient. Next, the analysis results showed that there were no significant differences in the number of central lymph node metastases, the number of retrieved lymph nodes in the lateral cervical region, and the number of metastases between RT and OT. Additionally, the number of retrieved lymph nodes in each group was significantly greater than the number of metastases, indicating that RT met the requirements of oncology safety. Finally, There was no statistically significant difference in the postoperative recurrence rate between the two groups. However, Heaton et al. showed that higher lymph node yield in central and lateral neck dissections is associated with lower rates of papillary thyroid cancer recurrence in the central and lateral neck [55]. Further studies are still needed.
Another noteworthy aspect was the cosmetic satisfaction rate after surgery for 3–6 months. The cosmetic satisfaction rate of patients in the RT group was significantly higher than that in the OT group. This is because the RT incision is in the axilla, which can be better hidden when the patient’s upper limb is in the natural state, while the OT neck incision is obvious, especially in patients with a scar constitution. This was the original intention for the design of endoscopic thyroid surgery.
With the rapid development of RT technology, the Food and Drug Administration approved the latest Da Vinci SP system, namely the single-hole Da Vinci Robot for clinical use in 2018, in which all joints and endoscopes were inserted through a single-arm chamber with two control joints. This type of robot can perform more complex fine operations in a smaller anatomical area, which may further improve the operation of RT and simplify tedious regulatory steps. Some scholars have conducted preliminary explorations [56].
This study systematically evaluated the efficacy of unilateral axillary approach robotic thyroid surgery compared with open thyroid surgery in the treatment of DTC but still had the following limitations: (1) The current research did not register on PROSPERO, although we strictly followed the steps of systematic review, there may still be bias [57]. (2) The 12 studies included in this meta-analysis were non-random or retrospective, and the criteria for temporary/permanent recurrent laryngeal nerve injury, temporary/permanent hypoparathyroidism, and hypocalcemia in each study may be different, although some studies have adopted the propensity score matching method, there may still exist bias. (3) The studies included originate all from the US and Korea, and some of these publications are authored by the same surgical teams, thus this meta-analysis may primarily reflect the experiences of a limited number of experts, potentially limiting the generalizability of the study results. (4) Most of the follow-up periods included in the literature were short. Since DTC is a relatively indolent malignant tumor, a longer follow-up period may be needed to evaluate the oncological safety of surgery.
5 Conclusion
Unilateral axillary robotic thyroid surgery may achieve results similar to those of traditional open surgery. Although the RT group had a longer total thyroidectomy time and more postoperative drainage, it showed shorter hospital stays, less intraoperative bleeding, less retrieved central lymph nodes, and higher cosmetic satisfaction, wherein all the differences were statistically significant (P < 0.05). To better understand RT, further validation in large prospective randomized controlled trials and long-term follow-up is required.
Supporting information
S1 Table. Systematic search strategy (PICOS strategy).
https://doi.org/10.1371/journal.pone.0298153.s001
(DOCX)
S2 Table. Up-to-date meta-analyses (including robotic trans-axillary approach).
https://doi.org/10.1371/journal.pone.0298153.s002
(DOCX)
Acknowledgments
We thank Professor Mingwei Wang for his valuable contributions to this research. We would also like to thank our families for their constant support and encouragement throughout this research.
References
- 1. Tae K, Ji YB, Song CM, Ryu J. Robotic and Endoscopic Thyroid Surgery: Evolution and Advances. Clinical and experimental otorhinolaryngology. 2019;12(1):1–11. pmid:30196688
- 2. Pan JH, Zhou H, Zhao XX, Ding H, Wei L, Qin L, et al. Robotic thyroidectomy versus conventional open thyroidectomy for thyroid cancer: a systematic review and meta-analysis. Surgical endoscopy. 2017;31(10):3985–4001. pmid:28337546
- 3. Sukato DC, Ballard DP, Abramowitz JM, Rosenfeld RM, Mlot S. Robotic versus conventional neck dissection: A systematic review and meta-analysis. Laryngoscope. 2019;129(7):1587–1596. pmid:30325513
- 4. Liu H, Wang Y, Wu C, Fei W, Luo E. Robotic surgery versus open surgery for thyroid neoplasms: a systematic review and meta-analysis. Journal of cancer research and clinical oncology. 2020;146(12):3297–3312. pmid:33108513
- 5. Lee J, Nah KY, Kim RM, Ahn YH, Soh EY, Chung WY. Differences in postoperative outcomes, function, and cosmesis: open versus robotic thyroidectomy. Surgical endoscopy. 2010;24(12):3186–3194. pmid:20490558
- 6. Kang SW, Lee SH, Park JH, Jeong JS, Park S, Lee CR, et al. A comparative study of the surgical outcomes of robotic and conventional open modified radical neck dissection for papillary thyroid carcinoma with lateral neck node metastasis. Surgical endoscopy. 2012;26(11):3251–3257. pmid:22648105
- 7. Lee J, Kwon IS, Bae EH, Chung WY. Comparative analysis of oncological outcomes and quality of life after robotic versus conventional open thyroidectomy with modified radical neck dissection in patients with papillary thyroid carcinoma and lateral neck node metastases. The Journal of clinical endocrinology and metabolism. 2013;98(7):2701–2708. pmid:23678034
- 8. Noureldine SI, Jackson NR, Tufano RP, Kandil E. A comparative North American experience of robotic thyroidectomy in a thyroid cancer population. Langenbeck’s archives of surgery. 2013;398(8):1069–1074. pmid:24057319
- 9. Ryu HR, Lee J, Park JH, Kang SW, Jeong JJ, Hong JY, et al. A comparison of postoperative pain after conventional open thyroidectomy and transaxillary single-incision robotic thyroidectomy: a prospective study. Annals of surgical oncology. 2013;20(7):2279–2284. pmid:23529779
- 10. Song CM, Ji YB, Bang HS, Park CW, Kim DS, Tae K. Quality of Life After Robotic Thyroidectomy by a Gasless Unilateral Axillary Approach. Annals of surgical oncology. 2014;21(13):4188–4194. pmid:25001095
- 11. Lee SG, Lee J, Kim MJ, Choi JB, Kim TH, Ban EJ, et al. Long-term oncologic outcome of robotic versus open total thyroidectomy in PTC: a case-matched retrospective study. Surgical endoscopy. 2016;30(8):3474–3479. pmid:26514137
- 12. Song CM, Ji YB, Sung ES, Kim DS, Koo HR, Tae K. Comparison of robotic versus conventional selective neck dissection and total thyroidectomy for papillary thyroid carcinoma. Otolaryngology—head and neck surgery: official journal of American Academy of Otolaryngology-Head and Neck Surgery. 2016;154(6):1005–1013. pmid:26980906
- 13. Tae K, Song CM, Ji YB, Sung ES, Jeong JH, Kim DS. Oncologic outcomes of robotic thyroidectomy: 5-year experience with propensity score matching. Surgical endoscopy. 2016;30(11):4785–4792. pmid:26895921
- 14. Kim MJ, Lee J, Lee SG, Choi JB, Kim TH, Ban EJ, et al. Transaxillary robotic modified radical neck dissection: a 5-year assessment of operative and oncologic outcomes. Surgical endoscopy. 2017;31(4):1599–1606. pmid:27572060
- 15. Garstka M, Mohsin K, Ali DB, Shalaby H, Ibraheem K, Farag M, et al. Well-differentiated thyroid cancer and robotic transaxillary surgery at a North American institution. The Journal of surgical research. 2018;228:170–178. pmid:29907208
- 16. Cho JW, Park JH, Lee YM, Hong SJ, Yoon JH. Comparison of short-term oncologic outcome of robotic thyroid surgery using dynamic risk stratification: A propensity score–matched comparison study. Surgery. 2019;165(3):608–616. pmid:30219245
- 17. Tae K, Ji YB, Jeong JH, Lee SH, Jeong MA, Park CW. Robotic thyroidectomy by a gasless unilateral axillo-breast or axillary approach: Our early experiences. Surgical endoscopy. 2011;25(1):221–228. pmid:20567849
- 18. Foley CS, Agcaoglu O, Siperstein AE, Berber E. Robotic transaxillary endocrine surgery: a comparison with conventional open technique. Surgical endoscopy. 2012;26(8):2259–2266. pmid:22311302
- 19. Landry CS, Grubbs EG, Warneke CL, Ormond M, Chua C, Lee JE, et al. Robot-assisted transaxillary thyroid surgery in the United States: Is it comparable to open thyroid lobectomy? Annals of surgical oncology. 2012;19(4):1269–1274. pmid:22065191
- 20. Lee J, Na KY, Kim RM, Oh Y, Lee JH, Lee J, et al. Postoperative functional voice changes after conventional open or robotic thyroidectomy: a prospective trial. Annals of surgical oncology. 2012;19(9):2963–2970. pmid:22535259
- 21. Lee S, Ryu HR, Park JH, Kim KH, Kang SW, Jeong JJ, et al. Early surgical outcomes comparison between robotic and conventional open thyroid surgery for papillary thyroid microcarcinoma. Surgery. 2012;151(5):724–730. pmid:22284763
- 22. Tae K, Ji YB, Cho SH, Lee SH, Kim DS, Kim TW. Early surgical outcomes of robotic thyroidectomy by a gasless unilateral axillo-breast or axillary approach for papillary thyroid carcinoma: 2 years’ experience. Head & neck. 2012;34(5):617–625. pmid:21688343
- 23. Tae K, Kim KY, Yun BR, Ji YB, Park CW, Kim DS, et al. Functional voice and swallowing outcomes after robotic thyroidectomy by a gasless unilateral axillo-breast approach: Comparison with open thyroidectomy. Surgical endoscopy. 2012;26(7):1871–1877. pmid:22205465
- 24. Aliyev S, Taskin HE, Agcaoglu O, Aksoy E, Milas M, Siperstein A, et al. Robotic transaxillary total thyroidectomy through a single axillary incision. Surgery. 2013;153(5):705–710. pmid:23294877
- 25. Yi O, Yoon JH, Lee YM, Sung TY, Chung KW, Kim TY, et al. Technical and oncologic safety of robotic thyroid surgery. Annals of surgical oncology. 2013;20(6):1927–1933. pmid:23306957
- 26. Lee S, Kim HY, Lee CR, Park S, Son H, Kang SW, et al. A prospective comparison of patient body image after robotic thyroidectomy and conventional open thyroidectomy in patients with papillary thyroid carcinoma. Surgery. 2014;156(1):117–25. pmid:24878453
- 27. Lee S, Lee CR, Lee SC, Park S, Kim HY, Son H, et al. Surgical completeness of robotic thyroidectomy: A prospective comparison with conventional open thyroidectomy in papillary thyroid carcinoma patients. Surgical endoscopy. 2014;28(9):1068–1075. pmid:24221277
- 28. Song CM, Ji YB, Bang HS, Park CW, Kim H, Tae K. Long-term sensory disturbance and discomfort after robotic thyroidectomy. World journal of surgery. 2014;38(7):1743–1748. pmid:24510245
- 29. Tae K, Song CM, Ji YB, Kim KR, Kim JY, Choi YY. Comparison of surgical completeness between robotic total thyroidectomy versus open thyroidectomy. The Laryngoscope. 2014;124(4):1042–1047. pmid:24338236
- 30. Song CM, Ji YB, Bang HS, Kim KR, Kim H, Tae K. Postoperative Pain after Robotic Thyroidectomy by a Gasless Unilateral Axillo-Breast or Axillary Approach. Surgical laparoscopy, endoscopy & percutaneous techniques. 2015;25(6):478–482. pmid:26422088
- 31. Arora A, Garas G, Sharma S, Muthuswamy K, Budge J, Palazzo F, et al. Comparing transaxillary robotic thyroidectomy with conventional surgery in a UK population: A case control study. International journal of surgery (London, England). 2016;27:110–117. pmid:26808320
- 32. Song CM, Yun BR, Ji YB, Sung ES, Kim KR, Tae K. Long-Term Voice Outcomes After Robotic Thyroidectomy. World journal of surgery. 2016;40(1):110–116. pmid:26464152
- 33. Sung TY, Yoon JH, Han M, Lee YH, Lee YM, Song DE, et al. Oncologic safety of robot thyroid surgery for papillary thyroid carcinoma: A Comparative Study of Robot versus Open Thyroid Surgery Using Inverse Probability of Treatment Weighting. PLoS ONE. 2016;11(6): e0157345. pmid:27285846
- 34. Fregoli L, Materazzi G, Miccoli M, Papini P, Guarino G, Wu HS, et al. Postoperative Pain Evaluation after Robotic Transaxillary Thyroidectomy Versus Conventional Thyroidectomy: a Prospective Study. Journal of laparoendoscopic & advanced surgical techniques. Part A. 2017;27(2):146–150. pmid:27875655
- 35. Lee IA, Kim K, Kim JK, Kang SW, Lee J, Jeong JJ, et al. Comparison of surgical outcomes between robotic transaxillary and conventional open thyroidectomy in pediatric thyroid cancer. Cancers. 2021;13(13):3293. pmid:34209221
- 36. Matteucci V, Bai D, Fregoli L, Papini P, Aghababyan A, Docimo G, et al. The effect of robot-assisted transaxillary thyroidectomy (RATT) on body image is better than the conventional approach with cervicotomy: a preliminary report. Updates in surgery. 2021;73(3):1169–1175. pmid:32399594
- 37. Rossi L, Buoni V, Fregoli L, Papini P, De Palma A, & Materazzi G. Postsurgical complications after robot-assisted transaxillary thyroidectomy: critical analysis of a large cohort of European patients. Updates in surgery. 2022;74(2):511–517. pmid:35239151
- 38. Tzelnick S, Tsur N, Amir I, Mizrachi A, Watt T, et al. Quality of Life Following Robotic Transaxillary Versus Conventional Hemithyroidectomy: A Comparative Analysis. Journal of laparoendoscopic & advanced surgical techniques. Part A. 2023;33(1):8–14. pmid:36322881
- 39. Kaliszewski K, Wojtczak B, Sutkowski K, Rudnicki J. Thyroid cancer surgery—in what direction are we going? A mini-review. The Journal of international medical research. 2020;48(4): 0300060520914803. pmid:32249645
- 40. Kim MJ, Nam KH, Lee SG, Choi JB, Kim TH, Lee CR, et al. Yonsei Experience of 5000 Gasless Transaxillary Robotic Thyroidectomies. World journal of surgery. 2018;42(2):393–401. pmid:28879559
- 41. Shen H, Shan C, Qiu M. Systematic review and meta-analysis of transaxillary robotic thyroidectomy versus open thyroidectomy. Surgical laparoscopy, endoscopy & percutaneous techniques. 2014;24(3):199–206. pmid:24887539
- 42. Wang YC, Liu K, Xiong JJ, Zhu JQ. Robotic thyroidectomy versus conventional open thyroidectomy for differentiated thyroid cancer: Meta-analysis. The Journal of laryngology and otology. 2015;129(6):558–567. pmid:26004423
- 43. Son SK, Kim JH, Bae JS, Lee SH, et al. Surgical Safety and Oncologic Effectiveness in Robotic versus Conventional Open Thyroidectomy in Thyroid Cancer: A Systematic Review and Meta-Analysis. Annals of surgical oncology. 2015;22(9):3022–3032. pmid:25634780
- 44. Sun GH, Peress L, Pynnonen MA. Systematic review and meta-analysis of robotic vs conventional thyroidectomy approaches for thyroid disease. Otolaryngology—head and neck surgery: official journal of American Academy of Otolaryngology-Head and Neck Surgery. 2014;150(4):520–532. pmid:24500878
- 45. Jackson NR, Yao L, Tufano RP, Kandil EH. Safety of robotic thyroidectomy approaches: meta-analysis and systematic review. Head & neck 2014;36(1):137–143. pmid:23471784
- 46. Liu SYW, Ng EKW. Robotic versus Open Thyroidectomy for Differentiated Thyroid Cancer: An Evidence-Based Review. International journal of endocrinology. 2016;2016:1–8. pmid:27069476
- 47. Lang BH, Wong CK, Tsang JS, Wong KP, Wan KY. A systematic review and meta-analysis evaluating completeness and outcomes of robotic thyroidectomy. The Laryngoscope. 2015;125(2):509–518. pmid:25236330
- 48. Kandil E, Hammad AY, Walvekar RR, Hu T, Masoodi H, Mohamed SE, et al. Robotic Thyroidectomy Versus Nonrobotic Approaches: A Meta-Analysis Examining Surgical Outcomes. Surgical innovation. 2016;23(3):317–325. pmid:26525401
- 49. D’Agostino RB Jr., Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Statistics in medicine. 1998;17(19):2265–2281. pmid:9802183
- 50. Rubin DB, Thomas N. Matching using estimated propensity scores: relating theory to practice. Biometrics. 1996;52(1):249–264. pmid:8934595
- 51. Kassite I, Bejan-Angoulvant T, Lardy H, Binet A. A systematic review of the learning curve in robotic surgery: range and heterogeneity. Surgical endoscopy. 2019;33(2):353–365. pmid:30267283
- 52. Song CM, Jang YI, Ji YB, Park JS, Kim DS, Tae K. Factors affecting operative time in robotic thyroidectomy. Head & neck. 2018;40(5):893–903. pmid:29206321
- 53. Sephton BM. Extracervical Approaches to Thyroid Surgery: Evolution and Review. Minimally invasive surgery. 2019;2019:5961690. pmid:31531238
- 54. Kim EB, Cho JW, Lee YM, Sung TY, Yoon JH, Chung KW, et al. Postsurgical Outcomes and Surgical Completeness of Robotic Thyroid Surgery: A Single Surgeon’s Experience on 700 Cases. Journal of laparoendoscopic & advanced surgical techniques. Part A. 2018;28(5):540–545. pmid:29346027
- 55. Heaton CM, Chang JL, Orloff L A. Prognostic Implications of Lymph Node Yield in Central and Lateral Neck Dissections for Well-Differentiated Papillary Thyroid Carcinoma. Thyroid, 2016; 26(3):434. pmid:26825749
- 56. Kim JK, Choi SH, Choi SM, Choi HR, Lee CR, Kang SW, et al. Single-port transaxillary robotic thyroidectomy (START): 200-cases with two-step retraction method. Surgical endoscopy. 2022;36(4):2688–2696. pmid:34741206
- 57. Cassai AD, Pasin L, Boscolo A, et al. Trial sequential analysis: plain and simple. Korean Society of Anesthesiologists. 2021;(4). pmid:33333708