Which is the best surgical approach for thymectomy: robot-assisted thoracoscopic surgery (RATS), video-assisted thoracoscopic surgery (VATS), thoracotomy (TORA) or subxiphoid video-assisted thoracoscopic surgery (SPT)?—a systematic review and network meta-analysis
Highlight box
Key findings
• We found that subxiphoid video-assisted thoracoscopic surgery thymectomy has faster recovery, less postoperative pain, and better perioperative efficacy than other surgical methods. Robot-assisted thoracoscopic surgery has high safety and certain clinical advantages.
What is known and what is new?
• Thymectomy has many surgical approaches, and all kinds of surgical approaches have their advantages and disadvantages. However, there is no clear conclusion which is the best surgical approach for thymectomy.
• We performed this network meta-analysis to compare the outcomes of different surgical approaches for thymectomy.
What is the implication, and what should change now?
• These findings may help clinicians on deciding the choice of proper surgical approaches for thymectomy.
Introduction
Thymectomy is indicated in the presence of primary thymic diseases such as thymomas, thymic cysts, suspicious anterior mediastinal lesions and myasthenia gravis (MG) (1,2). For benign lesions such as thymoma or thymic cyst, the main purpose of surgery is to remove the tumor so as to reduce its pressure on the surrounding tissues. Thymectomy is an effective treatment for MG, which is a rare autoimmune disease that affects neuromuscular transmission. Thymus plays a central role in the complex pathogenesis and persistence of this disease, involving the mechanisms of self tolerance and autoimmunity. Therefore, thymectomy can block these mechanisms, thereby improving or alleviating the symptoms of MG. Therefore, thymectomy is crucial for the treatment of thymic diseases (3).
Traditionally, the standard surgical treatment in the thymus pathology (both in benign and malignant cases) has been total thymectomy with sternotomy, which is a type of thoracotomy (TORA) (4). However, the TORA is invasive, and patients show a slow recovery. In recent years, with the development of thoracoscopy technology, the application of video-assisted thoracoscopic surgery (VATS) has resulted in a leap in the effectiveness of minimally invasive surgery and thereby in thymectomy. The minimally invasive approaches have become increasingly popular for the low intraoperative morbidity and mortality, improved anesthetic results, reduced access trauma and postoperative pain, shorter hospital stay and equivalent efficacy when compared with the conventional open techniques (5). With the increasing popularity of robotic surgery, robot-assisted thoracoscopic surgery (RATS) has gradually found its application in thymectomy. In comparison to conventional minimally invasive surgical instruments, Da Vinci system offers seven degrees of freedom, enabling precise operations within narrow anatomical areas. The Endo-Wrist system can articulate and rotate 360 degrees, thus improving maneuverability around anatomic structures (6). In addition, the Endo-Wrist system has various functions and eliminates the physiological vibrations of the hands of the surgeon, thereby reducing surgical risks and enhancing procedural safety (7). The traditional thoracoscopic surgery approach is the transthoracic intercostal approach, while the subxiphoid video-assisted thoracoscopic surgery (SPT) is also favored by many scholars due to its advantages of less pain and faster postoperative recovery (8). Therefore, there are currently various surgical methods for thymectomy, each with its own advantages and disadvantages. Clinicians need to select the appropriate surgical method based on different situations.
At present, there are many studies on the pairwise comparison of different surgical approaches for thymectomy, but there is still a lack of systematic studies on multiple surgical approaches.
Network meta-analysis (NMA), also known as multiple-treatments comparison (MTC), enables us to synthesize data from direct (within-trail) comparisons and can provide indirect (inter-trail) comparisons of multiple surgical approaches when direct comparisons are unavailable (9). In addition, the cluster analysis enables us to estimate the rank such as, which of the treatments is the best, the second best, etc. Thus, we performed a NMA to compare the perioperative data of different surgical approaches and to research which is the best surgical approach for thymectomy. We present this article in accordance with the PRISMA-NMA reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-24-443/rc).
Methods
Literature and database search strategy
We conducted a systematic literature search of PubMed, Excerpt Medica Database (EMBASE), Web of Science, and the Cochrane Central Register of Controlled Trials (CENTRAL) from its inception until May 1, 2024. We employed Medical Subject Headings/Emtree together with free text words like thymectomy, anterior mediastinal mass, anterior mediastinal tumor, VATS, RATS, SPT, open, sternotomy, video-assisted thoracoscopic, subxiphoid, and robot-assisted thoracoscopic. Furthermore, reference lists for eligible published clinical trials and meta-analyses were manually tracked to discover additional relevant studies. Only studies published in English were considered. Two reviewers (Q.Z. and Y.S.) independently evaluated the initially identified publications.
Selection criteria
Eligible studies were selected using the following inclusion criteria: (I) the study was a controlled trial; (II) many surgical methods were used; (III) the research subjects were patients with thymic diseases or anterior mediastinal tumors; (IV) the literature included perioperative outcomes such as operative time, blood loss, pleural drainage duration, pleural drainage volume, duration of hospital stay, intensive care unit (ICU) stay, visual analogue scale (VAS) score (postoperative day 1, POD1), postoperative complications, and survival data. Studies were excluded if the following criteria were met: (I) the articles were meta-analyses, letters, reviews, editorial materials, meeting abstracts, case reports and expert opinions; (II) not controlled trail; (III) not English literature; (IV) studies without adequate information. Two authors (Q.Z. and Y.S.) independently assessed the titles and abstracts of studies to identify whether these studies met the inclusion criteria. In the case of discrepancies, the two authors reached consensus via discussion.
Data extraction and quality assessment
Data were extracted from the selected studies by two independent investigators (Q.Z. and Y.S.). The following information were extracted: (I) publication data including first author, publication year, sample size and operative approaches; (II) the perioperative outcomes, including operative time, blood loss, pleural drainage duration, pleural drainage volume, duration of hospital stay, ICU stay, VAS score (POD1), postoperative complications, and survival data. Newcastle-Ottawa Quality Assessment Scale (NOS) was used to estimate the quality of every original study (10). For a semi-quantitative assessment, all three perspectives were considered: selection, comparability, and exposure and outcome ascertainment. A study was considered high-quality if its NOS was higher than 6 (11).
Statistical analysis
We calculated odds ratios (ORs) and their 95% confidence intervals (CIs) under either a fixed effect model or a random effect model as the relevant summarized statistics to evaluate the impact of various operative techniques for thymectomy. The Z-test was then run to determine whether the pooled effect size was significant (12).
The I2 test and the Cochran Q-statistic were used to assess the heterogeneity among the pooled studies (13). When there was considerable heterogeneity between studies (P<0.1 or I2>50%), the random effect model was employed. A fixed effect model was used otherwise.
A network evidence plot was created, where the width of the lines indicated the precision of the effect size comparison between two studies, the nodes represented interventions, and the node size represented sample size.
The efficacy ranks of various operational procedures were determined by comparing their surface under the cumulative ranking (SUCRA) values using a SUCRA curve. The higher the SUCRA value, the less effective the approach (14). To examine the effectiveness of several operational techniques based on the similarity of two variables, cluster analyses were used (14).
The small-study impact was assessed using a comparison-adjusted funnel plot, which took into account the variation in the overall effect for every group of studies (15).
Finally, we declared that STATA (version 18.0) was used to conduct all the statistical analyses listed above. All statistical tests were two-tailed, with a P<0.05 indicating statistical significance.
Results
Overview of the literature search
The study selection process is shown in Figure 1. Computer-based database searches and complementary manual search retrieved a total of 1,026 relevant articles. After removing 494 duplicates, we read the titles and abstracts of the 532 studies left, 199 studies were excluded because they either did not English studies (n=43), or were letters, reviews, meta-analyses, meeting abstracts and case reports (n=156). After meticulously reading, 275 studies were excluded because 249 studies were not controlled trials, and 26 studies were not related to surgery. In total, 58 eligible studies with 5,517 patients were enrolled in this NMA. The eligible studies were presented in Appendix 1.
Characteristics of the included studies
Fifty-eight articles published between 2005 and 2023, consisting of a total number of 5,517 participants were included in this NMA. There were 55 two-arm studies and three three-arm studies with four comparisons. Briefly, study sample sizes ranged from 21 to 288. Among all individuals, 797 were treated with RATS, 2,351 with VATS, 1,640 with TORA, and remaining 729 were treated with SPT. The detailed characteristics of the included studies were displayed in Table 1.
Table 1
| First author | Year | Patients | Interventions | Sample size | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RATS | VATS | TORA | SPT | Total | RATS | VATS | TORA | SPT | ||||
| Agatsuma H | 2017 | Thymomas | – | √ | √ | – | 280 | – | 140 | 140 | – | |
| Bagheri R | 2018 | – | – | √ | √ | – | 42 | – | 21 | 21 | – | |
| Balduyck B | 2011 | – | √ | – | √ | – | 36 | 14 | – | 22 | – | |
| Cao M | 2021 | Masaoka stage I–II thymomas | – | √ | – | √ | 84 | – | 51 | – | 33 | |
| Cao P | 2022 | Masaoka stage I–II thymomas | – | √ | – | √ | 137 | – | 72 | – | 65 | |
| Casiraghi M | 2018 | Masaoka stage I–II thymomas | √ | – | √ | – | 48 | 24 | – | 24 | – | |
| Chao YK | 2015 | Masaoka stage I–II thymomas | – | √ | √ | – | 96 | – | 48 | 48 | – | |
| Cheng YJ | 2005 | Masaoka stage II thymomas | – | √ | √ | – | 22 | – | 12 | 10 | – | |
| Derderian SC | 2020 | – | – | √ | √ | – | 34 | – | 16 | 18 | – | |
| Erşen E | 2018 | – | – | √ | √ | – | 40 | – | 17 | 23 | – | |
| Fiorelli A | 2016 | – | – | √ | √ | – | 43 | – | 23 | 20 | – | |
| Gu ZT | 2015 | Masaoka stage I–II thymomas | – | √ | √ | – | 93 | – | 49 | 44 | – | |
| Hajjar WM | 2020 | – | – | √ | √ | – | 50 | – | 30 | 20 | – | |
| He Z | 2013 | Masaoka stage I–II thymomas | – | √ | √ | – | 33 | – | 15 | 18 | – | |
| Imielski B | 2020 | – | √ | √ | √ | – | 220 | 54 | 97 | 69 | – | |
| Jiang L | 2022 | Masaoka stage I–II thymomas | – | √ | – | √ | 228 | – | 189 | – | 39 | |
| Jun Y | 2014 | Thymus | √ | √ | – | – | 115 | 55 | 60 | – | – | |
| Jurado J | 2012 | Masaoka stage I–III thymomas | – | √ | √ | – | 245 | – | 77 | 168 | – | |
| Kamel MK | 2017 | – | √ | √ | √ | – | 89 | 70 | 7 | 12 | – | |
| Kang CH | 2016 | Masaoka stage I -IV thymomas | √ | – | √ | – | 200 | 100 | – | 100 | – | |
| Kimura T | 2013 | Masaoka stage I–II thymomas | – | √ | √ | – | 74 | – | 45 | 29 | – | |
| Kneuertz PJ | 2017 | Masaoka stage I -IV thymomas | √ | – | √ | – | 54 | 20 | – | 34 | – | |
| Li XK | 2020 | Thymic epithelial tumors | √ | √ | – | – | 120 | 60 | 60 | – | – | |
| Li Y | 2023 | – | – | √ | – | √ | 114 | – | 57 | – | 57 | |
| Liu TJ | 2014 | – | – | √ | √ | – | 120 | – | 76 | 44 | – | |
| Liu Z | 2021 | – | – | √ | – | √ | 152 | – | 76 | – | 76 | |
| Lo CM | 2014 | – | – | √ | √ | – | 83 | – | 44 | 39 | – | |
| Lu Q | 2018 | Masaoka stage I–II thymomas | – | √ | – | √ | 77 | – | 36 | – | 41 | |
| Luzzi L | 2021 | – | √ | – | √ | – | 114 | 57 | – | 57 | – | |
| Maniscalco P | 2015 | Masaoka stage I–II thymomas | – | √ | √ | – | 27 | – | 13 | 14 | – | |
| Manoly I | 2014 | – | – | √ | √ | – | 39 | – | 17 | 22 | – | |
| Mao Y | 2020 | Anterior mediastinal tumor | – | √ | – | √ | 78 | – | 39 | – | 39 | |
| Marulli G | 2018 | Masaoka stage I–II thymomas | √ | – | √ | – | 164 | 56 | – | 108 | – | |
| Muhammad MI | 2014 | – | – | √ | √ | – | 21 | – | 13 | 8 | – | |
| Odaka M | 2010 | Masaoka stage I–II thymomas | – | √ | √ | – | 40 | – | 22 | 18 | – | |
| Odaka M | 2015 | Masaoka stage I–II thymomas | – | √ | √ | – | 98 | – | 67 | 31 | – | |
| Pennathur A | 2011 | Masaoka stage I–II thymomas | – | √ | √ | – | 40 | – | 18 | 22 | – | |
| Qian L | 2017 | Masaoka stage I–II thymomas | √ | √ | √ | – | 123 | 51 | 35 | 37 | – | |
| Qiu Z | 2020 | – | – | √ | – | √ | 109 | – | 51 | – | 58 | |
| Rowse PG | 2015 | Masaoka stage I–II thymomas | √ | √ | – | – | 56 | 11 | 45 | – | – | |
| Rückert JC | 2011 | – | √ | √ | – | – | 153 | 74 | 79 | – | – | |
| Sakamaki Y | 2014 | Masaoka stage I–II thymomas | – | √ | √ | – | 82 | – | 71 | 11 | – | |
| Şehitogullari A | 2020 | Masaoka stage I–II thymomas | √ | √ | – | – | 45 | 21 | 24 | – | – | |
| Seong YW | 2014 | – | √ | – | √ | – | 68 | 34 | – | 34 | – | |
| Siwachat S | 2018 | – | – | √ | √ | – | 98 | – | 53 | 45 | – | |
| Tagawa T | 2014 | Masaoka stage I–II thymomas | – | √ | √ | – | 27 | – | 15 | 12 | – | |
| Tang Y | 2015 | – | – | √ | – | √ | 45 | – | 25 | – | 20 | |
| Weksler B | 2012 | Masaoka stage I–III thymomas | √ | – | √ | – | 50 | 15 | – | 35 | – | |
| Wilshire CL | 2016 | Masaoka stage I–III thymomas | √ | – | √ | – | 40 | 23 | – | 17 | – | |
| Witte Pfister A | 2017 | – | √ | √ | – | – | 22 | 14 | 8 | – | – | |
| Xu H | 2020 | Masaoka stage I–II thymomas | – | √ | – | √ | 107 | – | 70 | – | 37 | |
| Yang X | 2023 | – | – | √ | – | √ | 288 | – | 144 | – | 144 | |
| Yano M | 2017 | – | – | √ | – | √ | 60 | – | 46 | – | 14 | |
| Ye B | 2013 | Masaoka stage I thymomas | √ | √ | – | – | 46 | 21 | 25 | – | – | |
| Ye B | 2014 Jan | Masaoka stage I–II thymomas | √ | – | √ | – | 74 | 23 | – | 51 | – | |
| Ye B | 2014 May | Masaoka stage I–II thymomas | – | √ | √ | – | 262 | – | 125 | 137 | – | |
| Yin X | 2023 | – | – | √ | √ | 156 | – | – | 78 | 78 | ||
| Zhang L | 2019 | Masaoka stage I–II thymomas | – | √ | – | √ | 56 | – | 28 | – | 28 | |
Study information is provided in Appendix 1. RATS, robot-assisted thoracoscopic surgery; SPT, subxiphoid video-assisted thoracoscopic surgery; TORA, thoracotomy; VATS, video-assisted thoracoscopic surgery.
Quality assessment
Two researchers were assigned to evaluate all of the included studies. The results of the quality assessment involving 58 studies were presented in Table 2. The mean NOS score was 8.379 (range from 7 to 9), which suggested a good quality level.
Table 2
| First author (year) | Selection | Comparability | Exposure | Total score | ||
|---|---|---|---|---|---|---|
| Assessment of outcome | Follow-up long enough for outcome | Adequacy of follow-up of cohorts | ||||
| Agatsuma H (2017) | 4 | 2 | 1 | 1 | 1 | 9 |
| Bagheri R (2018) | 4 | 2 | 1 | 1 | 1 | 9 |
| Balduyck B (2011) | 4 | 2 | 1 | 1 | 1 | 9 |
| Cao M (2021) | 4 | 2 | 1 | 1 | 1 | 9 |
| Cao P (2022) | 4 | 2 | 1 | 1 | 0 | 8 |
| Casiraghi M (2018) | 4 | 2 | 1 | 1 | 1 | 9 |
| Chao YK (2015) | 4 | 2 | 1 | 1 | 0 | 8 |
| Cheng YJ (2005) | 4 | 2 | 1 | 1 | 1 | 9 |
| Derderian SC (2020) | 4 | 2 | 1 | 1 | 1 | 9 |
| Erşen E (2018) | 4 | 2 | 1 | 1 | 1 | 9 |
| Fiorelli A (2016) | 4 | 2 | 1 | 1 | 0 | 8 |
| Gu ZT (2015) | 4 | 2 | 1 | 0 | 0 | 7 |
| Hajjar WM (2020) | 4 | 2 | 1 | 1 | 1 | 9 |
| He Z (2013) | 4 | 2 | 1 | 1 | 1 | 9 |
| Imielski B (2020) | 4 | 2 | 1 | 0 | 0 | 7 |
| Jiang L (2022) | 4 | 2 | 1 | 1 | 1 | 9 |
| Jun Y (2014) | 4 | 2 | 1 | 0 | 0 | 7 |
| Jurado J (2012) | 4 | 2 | 1 | 0 | 0 | 7 |
| Kamel MK (2017) | 4 | 2 | 1 | 1 | 1 | 9 |
| Kang CH (2016) | 4 | 2 | 1 | 1 | 1 | 9 |
| Kimura T (2013) | 4 | 2 | 1 | 1 | 0 | 8 |
| Kneuertz PJ (2017) | 4 | 2 | 1 | 1 | 1 | 9 |
| Li XK (2020) | 4 | 2 | 1 | 1 | 1 | 9 |
| Li Y (2023) | 4 | 2 | 1 | 1 | 1 | 9 |
| Liu TJ (2014) | 4 | 2 | 1 | 0 | 1 | 8 |
| Liu Z (2021) | 4 | 2 | 1 | 1 | 0 | 8 |
| Lo CM (2014) | 4 | 2 | 1 | 1 | 0 | 8 |
| Lu Q (2018) | 4 | 2 | 1 | 0 | 1 | 8 |
| Luzzi L (2021) | 4 | 2 | 1 | 1 | 1 | 9 |
| Maniscalco P (2015) | 4 | 2 | 1 | 0 | 1 | 8 |
| Manoly I (2014) | 4 | 2 | 1 | 1 | 0 | 8 |
| Mao Y (2020) | 4 | 2 | 1 | 1 | 1 | 9 |
| Marulli G (2018) | 4 | 2 | 1 | 1 | 0 | 8 |
| Muhammad MI (2014) | 4 | 2 | 1 | 1 | 1 | 9 |
| Odaka M (2010) | 4 | 2 | 1 | 0 | 0 | 7 |
| Odaka M (2015) | 4 | 2 | 1 | 1 | 0 | 8 |
| Pennathur A (2011) | 4 | 2 | 1 | 1 | 1 | 9 |
| Qian L (2017) | 4 | 2 | 1 | 0 | 0 | 8 |
| Qiu Z (2020) | 4 | 2 | 1 | 1 | 0 | 8 |
| Rowse PG (2015) | 4 | 2 | 1 | 0 | 1 | 8 |
| Rückert JC (2011) | 4 | 2 | 1 | 1 | 1 | 9 |
| Sakamaki Y (2014) | 4 | 2 | 1 | 1 | 1 | 9 |
| Şehitogullari A (2020) | 4 | 2 | 1 | 1 | 0 | 8 |
| Seong YW (2014) | 4 | 2 | 1 | 1 | 1 | 9 |
| Siwachat S (2018) | 4 | 2 | 1 | 0 | 1 | 8 |
| Tagawa T (2014) | 4 | 2 | 1 | 1 | 1 | 9 |
| Tang Y (2015) | 4 | 2 | 1 | 0 | 1 | 8 |
| Weksler B (2012) | 4 | 2 | 1 | 1 | 0 | 8 |
| Wilshire CL (2016) | 4 | 2 | 1 | 1 | 0 | 8 |
| Witte Pfister A (2017) | 4 | 2 | 1 | 1 | 1 | 9 |
| Xu H (2020) | 4 | 2 | 1 | 1 | 0 | 8 |
| Yang X (2023) | 4 | 2 | 1 | 0 | 1 | 8 |
| Yano M (2017) | 4 | 2 | 1 | 1 | 1 | 9 |
| Ye B (2013) | 4 | 2 | 1 | 0 | 1 | 8 |
| Ye B (2014 Jan) | 4 | 2 | 1 | 1 | 0 | 8 |
| Ye B (2014 May) | 4 | 2 | 1 | 1 | 1 | 9 |
| Yin X (2023) | 4 | 2 | 1 | 1 | 1 | 9 |
| Zhang L (2019) | 4 | 2 | 1 | 0 | 0 | 7 |
Study information is provided in Appendix 1.
Pairwise meta-analyses for perioperative outcomes of different surgical approaches
We conducted direct-paired comparisons of outcomes of different surgical approaches, and the results were displayed in Figure 2 and Table 3. The results demonstrated that RATS had less blood loss compared with VATS and TORA [−0.337 (−0.616, −0.059), P=0.02; −1.092 (−1.313, −0.871), P<0.001], VATS had less blood loss compared with TORA [−0.691 (−0.832, −0.550), P<0.001]. In terms of pleural drainage volume, VATS had less pleural drainage volume compared with TORA [−0.475 (−0.732, −0.218), P<0.001]. In terms of VAS score, VATS had greater VAS score compared with SPT [1.568 (1.434, 1.702), P<0.001].
Table 3
| Included studies | Comparisons | Sample size | Pairwise meta-analysis | ||||
|---|---|---|---|---|---|---|---|
| Operation 1 | Operation 2 | OR (95% CI) | I2 | P | |||
| Operative time (min) | |||||||
| 8 studies | RATS vs. VATS | 347 | 425 | 0.063 (−0.086–0.211) | 91.5% | 0.41 | |
| 8 studies | RATS vs. TORA | 357 | 394 | −0.145 (−0.296–0.005) | 94.1% | 0.058 | |
| 19 studies | VATS vs. TORA | 675 | 517 | −0.184 (−0.306 to −0.061)* | 90.5% | 0.003 | |
| 12 studies | VATS vs. SPT | 845 | 612 | −0.224 (−0.334 to −0.113)* | 92.7% | <0.001 | |
| 1 study | TORA vs. SPT | 78 | 78 | 1.971 (1.588–2.354)* | NA | NA | |
| Blood loss (mL) | |||||||
| 4 studies | RATS vs. VATS | 104 | 129 | −0.337 (−0.616 to −0.059)* | 84.3% | 0.02 | |
| 4 studies | RATS vs. TORA | 189 | 223 | −1.092 (−1.313 to −0.871)* | 95.9% | <0.001 | |
| 15 studies | VATS vs. TORA | 477 | 407 | −0.691 (−0.832 to −0.550)* | 84.2% | <0.001 | |
| 11 studies | VATS vs. SPT | 817 | 584 | 0.200 (0.087–0.312)* | 89.5% | <0.001 | |
| 1 study | TORA vs. SPT | 78 | 78 | 4.645 (4.039–5.252)* | NA | NA | |
| Pleural drainage duration (days) | |||||||
| 4 studies | RATS vs. VATS | 153 | 144 | −0.708 (−0.949 to −0.466)* | 88.3% | <0.001 | |
| 4 studies | RATS vs. TORA | 128 | 156 | −1.711 (−2.032 to −1.390)* | 97.8% | <0.001 | |
| 9 studies | VATS vs. TORA | 304 | 262 | −0.501 (−0.672 to −0.330)* | 67.5% | <0.001 | |
| 11 studies | VATS vs. SPT | 772 | 534 | 0.160 (0.042–0.278)* | 84.3% | 0.01 | |
| Pleural drainage volume (mL) | |||||||
| 3 studies | RATS vs. VATS | 132 | 119 | −0.663 (−0.936 to −0.389)* | 96.6% | <0.001 | |
| 3 studies | RATS vs. TORA | 108 | 122 | −2.669 (−3.066 to −2.272)* | 97.4% | <0.001 | |
| 5 studies | VATS vs. TORA | 139 | 142 | −0.475 (−0.732 to −0.218)* | 94.8% | <0.001 | |
| 6 studies | VATS vs. SPT | 531 | 370 | 0.003 (−0.136–0.142) | 0.0% | 0.96 | |
| Duration of hospital stay (days) | |||||||
| 5 studies | RATS vs. VATS | 207 | 241 | −0.648 (−0.846 to −0.450)* | 89.5% | <0.001 | |
| 7 studies | RATS vs. TORA | 257 | 294 | −1.146 (−1.337 to −0.956)* | 92.7% | <0.001 | |
| 19 studies | VATS vs. TORA | 651 | 555 | −0.786 (−0.907 to −0.666)* | 77.3% | <0.001 | |
| 12 studies | VATS vs. SPT | 827 | 594 | 0.321 (0.209–0.433)* | 88.8% | <0.001 | |
| Duration of intensive care unit (ICU) stay (days) | |||||||
| 1 study | RATS vs. VATS | 54 | 97 | −0.153 (−0.486–0.180) | NA | NA | |
| 2 studies | RATS vs. TORA | 111 | 126 | −0.471 (−0.730 to −0.211)* | 0.0% | <0.001 | |
| 9 studies | VATS vs. TORA | 399 | 324 | −0.686 (−0.845 to −0.528)* | 92.6% | <0.001 | |
| Visual analogue scale (VAS) score (POD1) | |||||||
| 1 study | VATS vs. TORA | 23 | 20 | −1.529 (−2.213 to −0.844)* | NA | NA | |
| 9 studies | VATS vs. SPT | 746 | 550 | 1.568 (1.434–1.702)* | 95.5% | <0.001 | |
| Postoperative complications | |||||||
| 7 studies | RATS vs. VATS | 252 | 229 | 1.530 (0.758–3.090)* | 0.0% | 0.24 | |
| 10 studies | RATS vs. TORA | 421 | 459 | 0.523 (0.340–0.805)* | 31.7% | 0.003 | |
| 15 studies | VATS vs. TORA | 734 | 791 | 0.590 (0.424–0.821)* | 0.0% | 0.002 | |
| 10 studies | VATS vs. SPT | 784 | 551 | 0.817 (0.557–1.199)* | 68.2% | 0.30 | |
*, statistically significant. CI, confidence interval; ICU, intensive care unit; NA, not available; OR, odds ratio; POD1, postoperative day 1; RATS, robot-assisted thoracoscopic surgery; SPT, subxiphoid video-assisted thoracoscopic surgery; TORA, thoracotomy; VAS, visual analogue scale; VATS, video-assisted thoracoscopic surgery.
Inconsistency tests for perioperative outcomes of different surgical approaches
Inconsistency tests showed that the results of direct and indirect evidences of different surgical approaches were consistency, the consistency model was adopted (all P>0.05) (Table 4). The network contribution graph reflected the weight of direct comparison (Figure 3).
Table 4
| Pairwise comparisons | Direct HR values | Indirect HR values | P value | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RATS vs. VATS | RATS vs. TORA | VATS vs. TORA | VATS vs. SPT | TORA vs. SPT | RATS vs. VATS | RATS vs. TORA | VATS vs. TORA | VATS vs. SPT | TORA vs. SPT | RATS vs. VATS | RATS vs. TORA | VATS vs. TORA | VATS vs. SPT | TORA vs. SPT | |||
| Operative time (min) | −2.51 | 21.13 | −1.27 | 9.07 | −64.10 | 26.56 | −1.26 | 33.17 | −62.35 | 7.32 | 0.19 | 0.32 | 0.10 | 0.06 | 0.06 | ||
| Blood loss (mL) | 16.44 | 243.52 | 83.38 | −8.08 | −94.00 | 176.95 | 82.89 | 183.94 | 6.42 | −108.58 | 0.25 | 0.47 | 0.07 | 0.83 | 0.83 | ||
| Pleural drainage duration (days) | 1.249 | 1.643 | 0.730 | −0.361 | NR | 0.925 | 2.113 | 0.314 | −2.305 | NR | 0.66 | 0.52 | 0.59 | >0.99 | NR | ||
| Pleural drainage volume (mL) | 124.601 | 401.743 | 251.154 | 14.463 | NR | 101.444 | 300.247 | 253.916 | −232.428 | NR | 0.90 | 0.59 | 0.99 | 0.99 | NR | ||
| Duration of hospital stay (days) | 1.421 | 3.444 | 2.698 | −1.165 | NR | 0.737 | 4.604 | 2.332 | −2.393 | NR | 0.58 | 0.36 | 0.80 | >0.99 | NR | ||
| Duration of ICU stay (days) | 0.190 | 1.568 | 0.898 | NR | NR | 1.303 | 1.030 | 5.800 | NR | NR | 0.46 | 0.77 | 0.19 | NR | NR | ||
| VAS score (POD1) | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | ||
| Postoperative complications | −0.276 | 0.628 | 0.601 | 0.241 | NR | −0.007 | 0.182 | 0.876 | 0.385 | NR | 0.67 | 0.48 | 0.66 | >0.99 | NR | ||
HR, hazard ratio; ICU, intensive care unit; NR, not reported; POD1, postoperative day 1; RATS, robot-assisted thoracoscopic surgery; SPT, subxiphoid video-assisted thoracoscopic surgery; TORA, thoracotomy; VAS, visual analogue scale; VATS, video-assisted thoracoscopic surgery.
Network meta-analyses for perioperative outcomes of different surgical approaches
The results of network meta-analyses in perioperative outcomes and the network evidence plots were displayed in Figure 4 and Table 5. The results suggested that in terms of blood loss, RATS had less blood loss than VATS, TORA and SPT [−77.81 (−129.81, −25.80); −177.25 (−229.55, −124.96); −70.92 (−133.80, −8.04); respectively], VATS had less blood loss than TORA [−99.45 (−132.07, −66.83)], TORA had more blood loss than SPT [106.34 (59.26, 153.41)]. As for pleural drainage volume, VATS had less pleural drainage volume than TORA [−566.45 (−1,033.65, −99.25)], TORA had more pleural drainage volume than SPT [551.85 (62.20, 1,041.50)]. In terms of VAS score, VATS and TORA had higher VAS score than SPT [2.02 (1.27, 2.76); 3.82 (1.39, 6.25); respectively]. However, the operative time, ICU stay and postoperative complications of all surgical approaches had no differences.
Table 5
| Perioperative outcomes | OR (95% CI) | |||
|---|---|---|---|---|
| RATS | VATS | TORA | SPT | |
| Operative time (min) | ||||
| vs. RATS | – | −25.06 (−84.50, 34.37) | 3.71 (−25.72, 33.14) | 10.64 (−49.56, 70.84) |
| vs. VATS | 25.06 (−34.37, 84.50) | – | 9.46 (−50.45, 69.37) | 9.07 (−11.89, 30.03) |
| vs. TORA | −3.71 (−33.14, 25.72) | −9.46 (−69.37, 50.45) | – | −0.39 (−63.85, 63.07) |
| vs. SPT | −10.64 (−70.84, 49.56) | −9.07 (−30.03, 11.89) | 0.39 (−63.07, 63.85) | – |
| Blood loss (mL) | ||||
| vs. RATS | – | 77.81 (25.80, 129.81)* | 177.25 (124.96, 229.55)* | 70.92 (8.04, 133.80)* |
| vs. VATS | −77.81 (−129.81, −25.80)* | – | 99.45 (66.83, 132.07)* | −6.89 (−43.58, 29.80) |
| vs. TORA | −177.25 (−229.55, −124.96)* | −99.45 (−132.07, −66.83)* | – | −106.34 (−153.41, −59.26)* |
| vs. SPT | −70.92 (−133.80, −8.04)* | 6.89 (−29.80, 43.58) | 106.34 (59.26, 153.41)* | – |
| Pleural drainage duration (days) | ||||
| vs. RATS | – | −0.50 (−2.52, 1.52) | 1.40 (0.32, 2.48) | 0.19 (−1.79, 2.17) |
| vs. VATS | 0.50 (−1.52, 2.52) | – | 0.10 (−1.92, 2.12) | −0.36 (−0.89, 0.16) |
| vs. TORA | −1.40 (−2.48, −0.32)* | −0.10 (−2.12, 1.92) | – | −0.46 (−2.55, 1.63) |
| vs. SPT | −0.19 (−2.17, 1.79) | 0.36 (−0.16, 0.89) | 0.46 (−1.63, 2.55) | – |
| Pleural drainage volume (mL) | ||||
| vs. RATS | – | 200.35 (−243.66, 644.35) | 61.35 (−181.38, 304.09) | −312.70 (−780.10, 154.70) |
| vs. VATS | −200.35 (−644.35, 243.66) | – | 566.45 (99.25, 1,033.65)* | 14.59 (−131.98, 161.17) |
| vs. TORA | −61.35 (−304.09, 181.38) | −566.45 (−1033.65, −99.25)* | – | −551.85 (−1,041.50, −62.20)* |
| vs. SPT | 312.70 (−154.70, 780.10) | −14.59 (−161.17, 131.98) | 551.85 (62.20, 1,041.50)* | – |
| Duration of hospital stay (days) | ||||
| vs. RATS | – | −0.46 (−3.58, 2.66) | 1.61 (−0.37, 3.60) | 1.27 (−1.76, 4.29) |
| vs. VATS | 0.46 (−2.66, 3.58) | – | 1.02 (−2.11, 4.16) | −1.17 (−2.22, −0.11) |
| vs. TORA | −1.61 (−3.60, 0.37) | −1.02 (−4.16, 2.11) | – | −2.19 (−5.50, 1.12) |
| vs. SPT | −1.27 (−4.29, 1.76) | 1.17 (0.11, 2.22) | 2.19 (−1.12, 5.50) | – |
| ICU stay (days) | ||||
| vs. RATS | – | 0.20 (−1.36, 1.76) | 2.45 (−1.39, 6.30) | – |
| vs. VATS | −0.20 (−1.76, 1.36) | – | 0.90 (−0.79, 2.59) | – |
| vs. TORA | −2.45 (−6.30, 1.39) | −0.90 (−2.59, 0.79) | – | – |
| VAS score (POD1) | ||||
| vs. VATS | – | – | 1.80 (−0.51, 4.11) | −2.02 (−2.76, −1.27)* |
| vs. TORA | – | −1.80 (−4.11, 0.51) | – | −3.82 (−6.25, −1.39)* |
| vs. SPT | – | 2.02 (1.27, 2.76)* | 3.82 (1.39, 6.25)* | – |
| Postoperative complications | ||||
| vs. RATS | – | 0.64 (−2.76, 4.04) | −0.39 (−1.33, 0.55) | −0.98 (−3.17, 1.21) |
| vs. VATS | −0.64 (−4.04, 2.76) | – | 1.91 (−0.38, 4.21) | 0.24 (−0.33, 0.81) |
| vs. TORA | 0.39 (−0.55, 1.33) | −1.91 (−4.21, 0.38) | – | −1.67 (−4.03, 0.69) |
| vs. SPT | 0.98 (−1.21, 3.17) | −0.24 (−0.81, 0.33) | 1.67 (−0.69, 4.03) | – |
*, statistically significant. CI, confidence interval; ICU, intensive care unit; OR, odds ratio; POD1, postoperative day 1; RATS, robot-assisted thoracoscopic surgery; SPT, subxiphoid video-assisted thoracoscopic surgery; TORA, thoracotomy; VAS, visual analogue scale; VATS, video-assisted thoracoscopic surgery.
Cumulative probability of perioperative outcomes of different surgical approaches
As shown in Figure 5 and Table 6, the SUCRA values of different surgical approaches demonstrated that in terms of operative time, the RATS ranked the lowest (19.5%), followed by VATS (46.9%). The blood loss of RATS ranked the lowest (1.2%), followed by SPT (43.0%). The pleural drainage duration of RATS ranked the lowest (1.8%), followed by SPT (34.5%). The pleural drainage volume of RATS ranked the lowest (11.1%), followed by VATS (44.0%). The duration of hospital stay of RATS ranked the lowest (17.2%), followed by SPT (17.7%). The ICU stay of RATS ranked the lowest (11.3%), followed by VATS (39.3%). The VAS score of SPT ranked the lowest (0.0%), followed by VATS (52.9%). The postoperative complications of VATS ranked the lowest (17.9%), followed by RATS (37.3%).
Table 6
| Surgical approach | SUCRA values (%) | |||||||
|---|---|---|---|---|---|---|---|---|
| Operative time (min) | Blood loss (mL) | Pleural drainage duration (days) | Pleural drainage volume (mL) | Duration of hospital stay (days) | ICU stay (days) | VAS score (POD1) | Postoperative complications | |
| RATS | 19.5 | 1.2 | 1.8 | 11.1 | 17.2 | 11.3 | NR | 37.3 |
| VATS | 46.9 | 55.8 | 64.1 | 44.0 | 65.1 | 39.3 | 52.9 | 17.9 |
| TORA | 72.1 | 100.0 | 99.6 | 99.7 | 100.0 | 99.4 | 97.1 | 94.4 |
| SPT | 61.5 | 43.0 | 34.5 | 45.2 | 17.7 | NR | 0.0 | 50.5 |
ICU, intensive care unit; NR, not reported; POD1, postoperative day 1; RATS, robot-assisted thoracoscopic surgery; SPT, subxiphoid video-assisted thoracoscopic surgery; SUCRA, surface under the cumulative ranking; TORA, thoracotomy; VAS, visual analogue scale; VATS, video-assisted thoracoscopic surgery.
Cluster analyses regarding perioperative outcomes in the included studies
The cluster analyses based on SUCRA values indicated that RATS and VATS had relatively lower operative time, blood loss, pleural drainage duration, pleural drainage volume, duration of hospital stay, ICU stay and postoperative complications, but the pain level of SPT was milder (Figure 6).
Publication bias regarding perioperative outcomes in the included studies
The comparison-adjusted funnel plots of perioperative outcomes of different surgical approaches showed that there were no publication bias among the included studies (Figure 7).
Remission of MG in different surgical approaches
We analyzed the improvement of clinical symptoms of MG after different surgical approaches, and counted the data of complete stable remission (CSR) and improvement respectively. The results showed that 15 articles reported relevant data, the MG related studies were presented in Appendix 2. The results of NMA suggested that there were no significant statistical difference in the remission of MG among the 4 different surgical approaches (Figure 8 and Table 7).
Table 7
| Remission of MG | OR (95% CI) | |||
|---|---|---|---|---|
| RATS | VATS | TORA | SPT | |
| CSR | ||||
| vs. RATS | – | −0.94 (−1.64, −0.23) | −1.03 (−1.85, −0.22) | −0.87 (−1.75, 0.01) |
| vs. VATS | 0.94 (0.23, 1.64) | – | −0.10 (−0.51, 0.32) | 0.06 (−0.46, 0.59) |
| vs. TORA | 1.03 (0.22, 1.85) | 0.10 (−0.32, 0.51) | – | 0.16 (−0.51, 0.83) |
| vs. SPT | 0.87 (−0.01, 1.75) | −0.06 (−0.59, 0.46) | −0.16 (−0.83, 0.51) | – |
| Improved | ||||
| vs. RATS | – | 0.37 (−2.53, 3.27) | 0.05 (−2.99, 3.10) | 0.09 (−0.75, 0.93) |
| vs. VATS | −0.37 (−3.27, 2.53) | – | −0.28 (−3.29, 2.74) | −0.07 (−0.55, 0.41) |
| vs. TORA | −0.05 (−3.10, 2.99) | 0.28 (−2.74, 3.29) | – | 0.20 (−2.85, 3.26) |
| vs. SPT | −0.09 (−0.93, 0.75) | 0.07 (−0.41, 0.55) | −0.20 (−3.26, 2.85) | – |
CI, confidence interval; CSR, complete stable remission; MG, myasthenia gravis; OR, odds ratio; RATS, robot-assisted thoracoscopic surgery; SPT, subxiphoid video-assisted thoracoscopic surgery; TORA, thoracotomy; VATS, video-assisted thoracoscopic surgery.
Discussion
The optimal approach for managing a thymic tumor or lesion entails the performance of a thymectomy (16). To some extent, tumor size affects the choice of surgical methods and the prognosis of tumors. Complete resection and invasive behavior at the pathological Masaoka-Koga stage are the most relevant and influential prognostic variables. In a study of 2,083 patients from the Japanese Association for Research of the Thymus (JART) database, Okumura et al. (17) discovered that a thymoma with a preoperative diameter greater than 5 cm was linked to a higher risk of recurrence and a lower 10-year relapse-free survival (RFS). Conversely, a larger thymoma with a diameter greater than 8 cm was linked to a lower disease-specific survival (DSS).
The TORA is traditionally used for thymectomy, which has good exposure, controllable operative risk and thorough resection, and is considered as the standard operation. With the popularity of thoracoscopic technology, its advantages of less trauma are gradually recognized. Initially, thoroughness of VATS was questioned. Many scholars conducted comparative studies on TORA and VATS, and found that VATS can achieve the same surgical effect as TORA (18,19), so it has gradually been widely recognized and applied in clinical practice. Thoracoscopic transthoracic approach may be relatively difficult to expose contralateral phrenic nerve and mediastinal fat, while SPT can fully expose bilateral phrenic nerve and mediastinal fat with less invasive surgery (8), which is favored by many clinicians. In recent years, RATS has been gradually applied in thymectomy, which is more flexible and more refined, making complex surgery more simple. Bongiolatti et al. discovered that RATS-thymectomy is a safe and effective treatment for treating patients with large thymomas (>5 cm), with comparable post-operative and long-term outcomes (20). To give valuable suggestions for different surgical approaches through comparing the perioperative outcomes, we conducted the NMAs among four surgical approaches commonly used for thymectomy, including RATS, VATS, TORA and SPT. The main advantage of our study over published systematic reviews is that we could compare a variety of surgical approaches simultaneously by applying the network method.
The results in the present study revealed no significant difference in surgical time among the four surgical approaches, which is different from the results of many studies (8,19,21). The possible reasons for this result include the following aspects: on the one hand, it may be due to different research period, resulting in different levels of surgical proficiency. On the other hand, it may be due to the presence of selective bias, with different studies incorporating varying degrees of surgical difficulty.
Focusing on the blood loss and pleural drainage volume, many studies have confirmed that thoracoscopic surgery is superior to TORA (22-24). Our study confirmed that the intraoperative blood loss and postoperative drainage volume of TORA were significantly increased compared to other surgical approaches, while the intraoperative blood loss of RATS was significantly lower than that of other surgical approaches. This is due to the larger surgical wound during TORA, RATS is precise and the wound is less prone to bleeding. However, there was no significant difference between the surgery under SPT and VATS, which is different from previous research (21). This maybe because both approaches are less invasive.
The SPT, which avoids intercostal compression and damage to intercostal nerves, significantly reduces postoperative pain. Additionally, it does not require one-lung ventilation or artificial pneumothorax, resulting in less respiratory system damage, faster recovery time, and shorter hospital stays (25). Several studies have found that by using the subxiphoid approach during surgery, the thymoma tissue, diagnosed as Masaoka III, invades the pericardium, lung tissue, and the left innominate vein can also completely be removed and reduce postoperative pain (26-30). This is consistent with our research findings.
Subsequently, the cluster analyses of perioperative outcomes of four surgical approaches demonstrated that SPT resulted in faster postoperative recovery, less postoperative pain and better quality of life, and other perioperative outcomes comparable to other surgical approaches. The RATS was safer and had certain clinical advantages.
The majority of prior trials on thymectomy for thymoma have revealed short-term surgical results but no oncologic outcomes (31). Pennathur et al. (32) compared 18 patients who received VATS thymectomy for stage I or II thymoma to 22 patients who got TORA thymectomy and discovered no significant differences in 5-year overall survival (OS) or RFS between the groups. Chung et al. (33) examined intermediate-term results in 25 individuals without MG who had VATS thymectomy and discovered 5- and 7-year RFS equivalent to those of TORA thymectomy. Furthermore, a multinational European investigation of the longest series of RATS thymectomy reported to date identified a low conversion rate to TORA and a 5-year OS comparable to TORA thymectomy (34). Long-term survival data were not assessed since there were none in this research.
We collected and analyzed the data related to the remission of MG through different surgical approaches, the results suggested that there were no significant statistical difference in the remission of MG among the four different surgical approaches. This result is similar to that of a previous study, which found that TORA thymectomy and RATS thymectomy both decreased myasthenic postoperative relapses and the average dose of steroids, but the disparity were not significant (18).
Several constraints should be highlighted. Firstly, there was a dearth of data on survival analysis in our included studies, more research is needed to assess the effectiveness of thymectomy. Secondly, the surgical level of various surgeons varied, which may have influenced the findings of our study. Thirdly, all of the data was obtained from published research, and no individual patient data were used. As a result, maintaining data quality was tough. Fourthly, the characteristics of the included papers varied, but our comparison-adjusted funnel plot did not reveal any asymmetry, indicating that there was no major publication bias. Fifthly, this study was lack of long-term survival data, which need further study in the future.
Conclusions
In conclusion, SPT has faster postoperative recovery, less postoperative pain, better quality of life, and other perioperative variables are not inferior to other surgical approaches. RATS is safer and has certain clinical advantages. TORA is gradually abandoned by surgeons, unless the case is difficult to manage under thoracoscopy, such as large tumors and invasion of adjacent vital organs. There were no significant statistical difference in the remission of MG among the four different surgical approaches. These findings may help clinicians on deciding the choice of proper surgical approaches for thymectomy.
Acknowledgments
We would like to thank the researchers and study participants for their contributions.
Footnote
Reporting Checklist: The authors have completed the PRISMA-NMA reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-24-443/rc
Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-24-443/prf
Funding: None.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-24-443/coif). The authors have no conflicts of interest to declare.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.
References
- Randomized Trial of Thymectomy in Myasthenia Gravis. N Engl J Med 2017;376:2097. [Crossref] [PubMed]
- Deeb ME, Brinster CJ, Kucharzuk J, et al. Expanded indications for transcervical thymectomy in the management of anterior mediastinal masses. Ann Thorac Surg 2001;72:208-11. [Crossref] [PubMed]
- Hussain K, Chen L, Gu Z, et al. Management of bleeding complications during thoracoscopic thymectomy. Mediastinum 2020;4:15. [Crossref] [PubMed]
- Cheng YJ, Kao EL, Chou SH. Videothoracoscopic resection of stage II thymoma: prospective comparison of the results between thoracoscopy and open methods. Chest 2005;128:3010-2. [Crossref] [PubMed]
- Hajjar WM, Almasry NW, Alarifi AS, et al. Comparison of the Outcomes Between Bilateral Thoracoscopic Maximal Thymectomy Versus Trans-Sternal Maximal Thymectomy for Non-Thyomomatous Myasthenia Gravis Patients: A Local University Hospital Experience. Cureus 2020;12:e12288. [Crossref] [PubMed]
- E H. Perioperative outcomes comparison of robotic and video-assisted thoracoscopic thymectomy for thymic epithelial tumor: a single-center experience. Updates Surg 2024;76:1511-9. [Crossref] [PubMed]
- Kajiwara N, Kakihana M, Kawate N, et al. Appropriate set-up of the da Vinci Surgical System in relation to the location of anterior and middle mediastinal tumors. Interact Cardiovasc Thorac Surg 2011;12:112-6. [Crossref] [PubMed]
- Jiang L, Chen H, Hou Z, et al. Subxiphoid Versus Unilateral Video-assisted Thoracoscopic Surgery Thymectomy for Thymomas: A Propensity Score Matching Analysis. Ann Thorac Surg 2022;113:1656-62. [Crossref] [PubMed]
- Thijs V, Lemmens R, Fieuws S. Network meta-analysis: simultaneous meta-analysis of common antiplatelet regimens after transient ischaemic attack or stroke. Eur Heart J 2008;29:1086-92. [Crossref] [PubMed]
- Lo CK, Mertz D, Loeb M. Newcastle-Ottawa Scale: comparing reviewers' to authors' assessments. BMC Med Res Methodol 2014;14:45. [Crossref] [PubMed]
- Wong WC, Cheung CS, Hart GJ. Development of a quality assessment tool for systematic reviews of observational studies (QATSO) of HIV prevalence in men having sex with men and associated risk behaviours. Emerg Themes Epidemiol 2008;5:23. [Crossref] [PubMed]
- Chen H, Manning AK, Dupuis J. A method of moments estimator for random effect multivariate meta-analysis. Biometrics 2012;68:1278-84. [Crossref] [PubMed]
- Jackson D, White IR, Riley RD. Quantifying the impact of between-study heterogeneity in multivariate meta-analyses. Stat Med 2012;31:3805-20. [Crossref] [PubMed]
- Chaimani A, Higgins JP, Mavridis D, et al. Graphical tools for network meta-analysis in STATA. PLoS One 2013;8:e76654. [Crossref] [PubMed]
- Peters JL, Sutton AJ, Jones DR, et al. Assessing publication bias in meta-analyses in the presence of between-study heterogeneity. J R Stat Soc Ser 2010;173:575-91.
- Wilshire CL, Vallières E, Shultz D, et al. Robotic Resection of 3 cm and Larger Thymomas Is Associated With Low Perioperative Morbidity and Mortality. Innovations (Phila) 2016;11:321-6. [Crossref] [PubMed]
- Okumura M, Yoshino I, Yano M, et al. Tumour size determines both recurrence-free survival and disease-specific survival after surgical treatment for thymoma. Eur J Cardiothorac Surg 2019;56:174-81. [Crossref] [PubMed]
- Lo CM, Lu HI, Hsieh MJ, et al. Thymectomy for myasthenia gravis: video-assisted versus transsternal. J Formos Med Assoc 2014;113:722-6. [Crossref] [PubMed]
- Friedant AJ, Handorf EA, Su S, et al. Minimally Invasive versus Open Thymectomy for Thymic Malignancies: Systematic Review and Meta-Analysis. J Thorac Oncol 2016;11:30-8. [Crossref] [PubMed]
- Bongiolatti S, Salvicchi A, Puzhlyiakov V, et al. Long-term outcomes of robot-assisted radical thymectomy for large thymomas: A propensity matched analysis. Int J Med Robot 2022;18:e2439. [Crossref] [PubMed]
- Wang CQ, Wang J, Liu FY, et al. Robot-assisted thoracoscopic surgery vs. sternotomy for thymectomy: A systematic review and meta-analysis. Front Surg 2022;9:1048547. [Crossref] [PubMed]
- Fiorelli A, Mazzella A, Cascone R, et al. Bilateral thoracoscopic extended thymectomy versus sternotomy. Asian Cardiovasc Thorac Ann 2016;24:555-61. [Crossref] [PubMed]
- Liu TJ, Lin MW, Hsieh MS, et al. Video-assisted thoracoscopic surgical thymectomy to treat early thymoma: a comparison with the conventional transsternal approach. Ann Surg Oncol 2014;21:322-8. [Crossref] [PubMed]
- Seong YW, Kang CH, Choi JW, et al. Early clinical outcomes of robot-assisted surgery for anterior mediastinal mass: its superiority over a conventional sternotomy approach evaluated by propensity score matching. Eur J Cardiothorac Surg 2014;45:e68-73; discussion e73. [Crossref] [PubMed]
- Li M, Xu L, Li L, et al. The Early Perioperative Outcomes of Subxiphoid Approach Versus Lateral Intercostal Approach Thoracoscopic Thymectomy for Thymic Tumors: A Meta-Analysis. J Laparoendosc Adv Surg Tech A 2022;32:256-64. [Crossref] [PubMed]
- Park SY, Han KN, Hong JI, et al. Subxiphoid approach for robotic single-site-assisted thymectomy. Eur J Cardiothorac Surg 2020;58:i34-8. [Crossref] [PubMed]
- Chen X, Ma Q, Wang X, et al. Subxiphoid and subcostal thoracoscopic surgical approach for thymectomy. Surg Endosc 2021;35:5239-46. [Crossref] [PubMed]
- Aramini B, Song N, Banchelli F, et al. Subxiphoid thymectomy with a double sternum retractor: a pilot study. Gland Surg 2019;8:657-62. [Crossref] [PubMed]
- Karunanantham J, Fok M, Ali JM, et al. Subxiphoid single incision thoracoscopic surgery approach for thymectomy: a case report. J Vis Surg 2017;3:147. [Crossref] [PubMed]
- Zieliński M, Rybak M, Wilkojc M, et al. Subxiphoid video-assisted thorascopic thymectomy for thymoma. Ann Cardiothorac Surg 2015;4:564-6. [Crossref] [PubMed]
- Davenport E, Malthaner RA. The role of surgery in the management of thymoma: a systematic review. Ann Thorac Surg 2008;86:673-84. [Crossref] [PubMed]
- Pennathur A, Qureshi I, Schuchert MJ, et al. Comparison of surgical techniques for early-stage thymoma: feasibility of minimally invasive thymectomy and comparison with open resection. J Thorac Cardiovasc Surg 2011;141:694-701. [Crossref] [PubMed]
- Chung JW, Kim HR, Kim DK, et al. Long-term results of thoracoscopic thymectomy for thymoma without myasthenia gravis. J Int Med Res 2012;40:1973-81. [Crossref] [PubMed]
- Marulli G, Rea F, Melfi F, et al. Robot-aided thoracoscopic thymectomy for early-stage thymoma: a multicenter European study. J Thorac Cardiovasc Surg 2012;144:1125-30. [Crossref] [PubMed]
