Check for updates
REVIEW ARTICLE - ENDOCRINE TUMORS
Open Versus Laparoscopic Surgery in the Management of Adrenocortical Carcinoma: A Systematic Review and Meta- analysis
Hayato Nakanishi, MSc1,2, Shahid Miangul, BSc1,2, Rongzhi Wang, MD3, Joe El Haddad, BSc1,2, Nour El Ghazal, BSc1,2, Fatma A. Abdulsalam, BSc1,2, Reem H. Matar, BSc1,2,4, Christian A. Than, PhD1,2,5, Benjamin E. Johnson, MD6, and Herbert Chen, MD3
1St George’s University of London, London, UK; 2University of Nicosia Medical School, University of Nicosia, Nicosia, Cyprus; 3Department of Surgery, University of Alabama at Birmingham, Birmingham, AL; 4Department of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN; 5School of Biomedical Sciences, The University of Queensland, St Lucia, QLD, Australia; 6Division of Surgery, NorthShore University Health System, Chicago, IL
ABSTRACT
Background. Laparoscopic surgery is considered a stan- dard treatment for benign adrenal tumors; however, no consensus has been reached on the optimal resection technique for adrenocortical carcinomas. This study aims to evaluate the safety and efficacy of laparoscopic surgery and open surgery in the management of adrenocortical carcinoma.
Methods. The Cochrane, Embase, PubMed, Scopus, and Web of Science databases were searched for articles from inception to May 2022, by two independent reviewers using the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines. The review was registered prospectively on the PROSPERO database (CRD42022316050).
Results. From 183 studies screened, 11 studies met the eligibility criteria, with a total of 1617 patients with adrenocortical carcinoma undergoing either laparoscopic surgery (n = 472) or open surgery (n = 1145). Open sur- gery demonstrated a lower rate of positive resection margin compared with laparoscopic surgery (odds ratio [OR] 1.52, 95% confidence interval [CI] 1.10-2.10; 12 = 0%). Addi- tionally, open surgery had more favorable overall survival
(OR 0.56, 95% CI 0.44-0.72; 12 = 0%) and recurrence-free rates (OR 0.60, 95% CI 0.42-0.85; 12 = 38%) than laparoscopic surgery at 3 years. Hospital stay was shorter for laparoscopic surgery than open surgery (mean differ- ence - 2.49 days, 95% CI - 2.95 to - 2.04; I 2 = 45%). Conclusions. Open surgery should still be considered the standard operative approach; however, laparoscopic sur- gery could be regarded as an effective and safe operation for selected adrenocortical carcinoma cases with appro- priate laparoscopic expertise. Further randomized controlled studies with tumor stage- and resection margin- dependent survival analysis are necessary to ascertain the safety and efficacy of the treatment.
Adrenocortical carcinoma (ACC) is a rare but aggres- sive malignancy with an annual incidence of 1.02 per million in the US.1 Depending on the tumor stage, the prognosis is generally poor, with 5-year survival rates reported between 15 and 44%.2 Given the unproven role of adjuvant therapy, including radiation and chemotherapy, margin-free surgical resection is regarded as the curative therapy of choice for ACC.
Open surgery (OS) has traditionally been the preferred surgical technique for ACC due to its ability to create better visual exposure of the tumor masses. For more advanced tumor stages, OS may require potential en bloc resection of the tumor and regionally infiltrated structures, which can involve lymphadenectomy or resection of the surrounding organs.3,4 In doing so, surgeons aim to
@ Society of Surgical Oncology 2022
Published Online: 7 November 2022
decrease the chance of perioperative complications, tumor capsule rupture, and, most critically, the chance of tumor recurrence by maintaining margin-free resection.3
Since its induction in 1992, laparoscopic surgery (LS) has become the surgical modality of choice for benign functioning and non-functioning adrenal disorders.5,6 Additionally, LS has proved effective in reducing the rate of perioperative morbidity, perioperative complication rates, cost reduction, and improved postoperative recovery, which have been well-documented in the literature.7- Despite the advantages of laparoscopic resections in benign adrenal tumors, its application in ACC remains a topic of controversy and debate. The primary reason is the higher rate at which local and port-site tumor recurrence have been reported in LS compared with OS.10,11 LS has also seen a higher risk of tumor capsule violation, which can elicit further complications such as peritoneal carcino- matosis.12 While this skepticism remains, increasing expertise in the field of LS has demonstrated its promising role in the preliminary exploration of suspected ACC and has shown similar oncological outcomes to OS in localized and small malignant masses. 13,14
While the role of LS in curative-intent therapy remains debatable, questions regarding short- and long-term effi- cacy, as well as complications, of LS compared with OS remain unanswered. To our knowledge, no meta-analysis has investigated the overall survival rate and recurrence- free rate based on time interval, as well as comparisons of negative margin survival rates, between LS and OS. Thus, this study aims to assess the safety and efficacy of LS compared with OS in the management of ACC.
METHODS
Data Sources and Search Strategies
A comprehensive search of several databases from each database’s inception to 6 May 2022 was conducted in compliance with the Preferred Reporting Items for Sys- tematic Reviews and Meta-analyses (PRISMA) guidelines.15 The databases included Ovid MEDLINE® and Epub Ahead of Print, In-Process and Other Non- Indexed Citations, Daily, Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, and Scopus. The search strategy was designed and conducted by an experienced librarian, with input from the study’s principal investigator. Controlled vocabulary supplemented with keywords was searched for comparative studies of minimally invasive surgery versus OS in the management of adult patients with ACC. The actual strategy listing all search terms used and how they are combined is available in electronic
supplementary material (ESM) Item 1. The review was registered prospectively in the PROSPERO database (CRD42022316050).
Eligibility Criteria and Quality Assessment
Eligible studies were cohort studies that met all the following inclusion criteria: (1) comparative studies of adult participants ≥ 18 years of age who underwent either OS or LS for the treatment of ACC; (2) outcomes of overall survival rates at 1, 2, 3, 5, or 10 years; and/or (3) outcomes of recurrence-free rates at 1, 2, 3, 5, or 10 years. Case reports, case series, abstracts, conference abstracts, and articles that were not reported in English were excluded from the study. This meta-analysis also excluded studies if a sample size was fewer than 10 patients in either the LS or OS group. Additionally, patients undergoing robotic adrenalectomy were excluded from the study. The quality of each study was independently evaluated by two authors (HN and SM) using the Newcastle-Ottawa Scale.16 Results of the quality assessment of all included studies are shown in ESM Item 2.
European Network for the Study of Adrenal Tumors (ENSAT) Stage Classification and Surgical Techniques
ACC tumor stage was designated in accordance with the American Joint Committee on Cancer/European Network for the Study of Adrenal Tumors (AJCC/ENSAT) guide- line: stage I, a tumor diameter ≤ 5 cm; stage II, a tumor diameter > 5 cm; stage III, tumor infiltration of neigh- boring structures, venous tumor thrombus in vena cava or vena renalis, or positive lymph nodes; and stage IV, distant metastases.17,18 Additionally, pathological examination for macro- or microscopic involvement of surgical margins was assessed for the surgical specimen. R classification was assigned based on the following categories: R0, no residual tumor; R1, microscopic residual tumor; R2, macroscopic residual tumor; and Rx, presence of residual tumor cannot be assessed.19 Lastly, patients who under- went conversion from laparoscopic to open were treated as the LS group based on intention-to-treat analysis.
Data Collection
1-, 2-, 3-, 5-, and 10-year survival rates were extracted and calculated from either reported literature values or raw data. When survival rates could not be directly obtained from the publication, Kaplan-Meier curves were digitized using WebPlotDigitizer version 4.4 (https://automeris.io/ WebPlotDigitizer/) and iteratively computed to generate individual patient data and survival rates using the algo- rithm from Guyot and colleagues.20 Furthermore, the
hazard ratio (HR) with a 95% confidence interval (CI) was analyzed based on the raw data obtained from Kaplan- Meier curves in the previous step and extracted to calculate logHR and standard error, as explained by Guyot and colleagues.2º These analyses were completed using R stu- dio software version 1.4.1106 (RStudio Team [2020]. RStudio: Integrated Development for R. RStudio, PBC, Boston, MA, USA).
Statistical Analysis
The pooled means and proportions of our data were analyzed using an inverse variance method for continuous data and the Mantel-Haenszel method for dichotomous data, which assigns the weight of each study based on its variance. The pooled estimate of the HR was analyzed using an inverse variance-weighted average of the individual studies.21 A direct comparison between the two techniques was conducted by assessing studies that reported outcomes of both treatments (two-arm analysis). The heterogeneity of effect size estimates across the studies was quantified using the Q statistic and 12. An 12 value of 0-25% indicates insignificant statistical heterogeneity, 26-50% indicates low heterogeneity, and 51-100% indicates high hetero- geneity.22 The random-effects model was used when the /2 value was > 50%, and the fixed-effects model was used for 12 < 50%. Furthermore, a leave-one-out sensitivity analysis was conducted to assess each study’s influence on the pooled estimate by omitting one study at a time and recal- culating the combined estimates for the remaining studies. Publication bias was assessed using a funnel plot.23 If the mean and standard deviation (SD) were unavailable, the median was converted to the mean using the formulas from the Cochrane Handbook for Systematic Reviews of Inter- ventions.21 Additionally, if the mean and SD were only depicted in figures, the mean and SD were digitized from figures using WebPlotDigitizer version 4.4 (https://autome ris.io/WebPlotDigitizer/). If the SD was not available or extractable, the reported mean was omitted from the cal- culation. Lastly, if a mixture of log-rank and Cox models are obtained from studies, all results were combined using the generic inverse variance method.21 Data analysis was performed using RevMan software version 5.4 (Review Manager (RevMan) [Computer program]. The Cochrane Collaboration, 2020, Copenhagen, Denmark).
RESULTS
Study Selection and Patient Characteristics
The initial literature search of the electronic databases yielded 183 studies. After title and abstract screening, 44
articles remained for full-text review. Those articles were then assessed for eligibility using specified inclusion and exclusion criteria. Finally, 11 studies (n = 1617) met the eligibility criteria and were included. Of the 11 studies, 424-27 7 were conducted at a multicenter, while 728-34 were performed at a single center. All the included studies were retrospective. The mean age of patients in the included studies ranged from 43.7 to 55.0 years, of whom 1005 (62.2 %) patients were female. A PRISMA flowchart of the study selection process is depicted in ESM Item 3. The baseline characteristics of the included studies are com- prehensively described in Table 1.
Risk of Bias
Results of the quality assessment of all the included studies are shown in ESM Item 2. Seven stud- ies 24,25,27,29,31,33,34 were judged to be of good quality. The patients appeared to represent the whole experience of the investigator. The exposure and outcome were adequately ascertained, and the lengths of follow-up were adequate. Conversely, four studies 26,28,30,32 were determined to be of poor quality. Those studies appeared to have a critical problem in the outcome/exposure domain due to significant loss to follow-up and an insufficient follow-up period. Nonetheless, all the included studies were considered adequate within selection and comparability domains.
Clinical Characteristics
Among the 11 included studies, a total of 1617 patients underwent either LS (n = 472) or OS (n = 1145); 561 (68.6%) and 763 (74.1%) patients in the LS and OS groups, respectively, were female. The mean length of the follow-up period was 48.1 ± 31.2 months for the LS cohort and 39.6 ± 31.1 months for the OS cohort. Among the seven included studies, 24,26-28,30,31,33 209 patients (53%) and 493 patients (47%) in the LS and OS groups, respectively, had ACC on the right side. Additionally, the transperitoneal surgical approach was used in 131 (91%) and 77 (79%) patients in the LS and OS groups, respectively. Nine stud- ies24,25,27,29-34 reported AJCC/ENSAT stage. Of these studies, the distribution of AJCC/ENSAT for stages I, II, and III was 26% (n = 43), 65% (n = 106), and 8% (n = 15) for LS, and 15% (n = 40), 74% (n = 199), and 10% (n = 30) for OS, respectively. Furthermore, tumor size was evaluated between the two groups, and the pooled estimate of tumor size demonstrated that the LS group was smaller than the OS group (mean difference [MD] - 2.92 cm, 95% CI - 4.02 to - 1.83; I 2 = 84%). In contrast, the pooled estimate of tumors being initially treated as incidentaloma (odds ratio [OR] 2.13, 95% CI 0.76-5.93; 12 = 58%) and Weiss score (MD 0.12, 95% CI - 0.52 to 0.77; I 2 = 86%) were
| Study | Publication year | Country | Study type | Total no. of participants (N) | Female sex [n (%)] | AJCC/ENSAT stage | Age, years (mean ± SD) | Length of follow-up, months (mean ± SD) | NOS scores (n) |
|---|---|---|---|---|---|---|---|---|---|
| Brix et al.24 | 2010 | Germany | Retrospective | 70 | 49 (70.0) | I-III | LS: 50.7 ± 13.8 | LS: 64 ± 22 | 6 |
| OS: 52.7 ± 16.8 | OS: 32 ± 31 | ||||||||
| Cooper et al. 28 | 2013 | USA | Retrospective | 302 | 196 (64.9) | I-IV | LS: 45.8 ± 13.9 | LS: 29.2 ± NR | 5 |
| OS: 46.5 ± 19.1 | OS: 35.5 ± NR | ||||||||
| Donatini et al. 29 | 2014 | France | Retrospective | 34 | 26 (76.5) | I-II | LS: 46 ± 14 | LS: 80 ± 32 | 7 |
| OS: 44 ± 19 | OS: 57 ± 33 | ||||||||
| Fossa et al. | 2013 | Norway | Retrospective | 32 | 23 (71.9) | I-III | LS: 45 ± 12 | LS: NR | 5 |
| OS: 52 ± 8 | OS: NR | ||||||||
| Kastelan et al. 25 | 2020 | Croatia | Retrospective | 46 | 32 (69.6) | I-III | LS: 43 ± 14 | LS: 53 ± 31 | 7 |
| OS: 51 ± 12 | OS: 51 ± 41 | ||||||||
| Lombardi et al.27 | 2012 | Italy | Retrospective | 156 | 100 (64.1) | I-II | LS: 52.0 ± 17.0 | LS: 50 ± 37 | 7 |
| OS: 46.6 ± 15.1 | OS: 40 ± 34 | ||||||||
| Mir et al.31 | 2013 | USA | Retrospective | 44 | 22 (50.0) | I-IV | LS: 53 ± 4 | LS: 18 ± 2 | 6 |
| OS: 48 ± 3 | OS: 31 ± 2 | ||||||||
| Porpiglia et al. 32 | 2010 | Italy | Retrospective | 43 | 26 (60.5) | I-II | LS: 47.0 ± 10.3 | LS: 30 ± 11 | 5 |
| OS: 41.3 ± 11.0 | OS: 38 ± 15 | ||||||||
| Tseng et al. | 2022 | USA | Retrospective | 804 | 481 (59.8) | I-III | LS: 56 ± 5 | LS: NR | 5 |
| OS: 55 ± 6 | OS: NR | ||||||||
| Wu et al. 33 | 2018 | China | Retrospective | 44 | 27 (61.4) | I-II | LS: 44.9 ± 16.2 | LS: 35 ± 25 | 6 |
| OS: 45.6 ± 14.5 | OS: 33 ± 24 | ||||||||
| Zheng et al. | 2018 | China | Retrospective | 42 | 23 (54.8) | I-III | LS: 45.2 ± 10.9 | LS: NR | 6 |
| OS: 48.5 ± 14.2 | OS: NR |
AJCC American Joint Committee of Cancer, ENSAT European Network for the Study of Adrenal Tumors, LS laparoscopic surgery, N number of patients, NOS Newcastle-Ottawa Scale, NR not reported, OS open surgery, SD standard deviation
| (a) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| LS | OS | Odds Ratio | Odds | Ratio | ||||||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-1 | I, Random, 95% CI | M-H, Random, | 95% CI | |||
| Brix 2010 | 8 | 35 | 9 | 35 | 29.6% | 0.86 [0.29, 2.56] | ||||||
| Fossa 2013 | 4 | 17 | 1 | 15 | 13.6% | 4.31 [0.42, 43.73] | ||||||
| Lombardi 2012 | 25 | 30 | 59 | 126 | 31.0% | 5.68 [2.04, 15.78] | ||||||
| Porpiglia 2010 | 6 | 18 | 7 | 25 | 25.8% | 1.29 [0.35, 4.78] | ||||||
| Total (95% CI) | 100 | 201 | 100.0% | 2.13 [0.76, 5.93] | ||||||||
| Total events | 43 | 76 | ||||||||||
| Heterogeneity: Tau2 = 0.61; Chi2 | = 7.19, | df = 3 | (P = | 0.07); I2 = | 58% | 0.01 | 0.1 1 | 10 | 100 | |||
| Test for overall effect: Z = 1.44 (P = 0.15) | Favours [LS] | Favours [OS] | ||||||||||
| Study or Subgroup | Mean | LS | Total | Mean | OS | Total | Weight | Mean Difference IV, Fixed, 95% CI | Mean Difference IV, Fixed, 95% CI |
|---|---|---|---|---|---|---|---|---|---|
| SD | SD | ||||||||
| Brix 2010 | 10 | 12.25 | 35 | 10 | 14.75 | 35 | 19.4% | 0.00 [-6.35, 6.35] | |
| Kastelan 2020 | 10 | 6 | 23 | 16 | 16 | 23 | 16.1% | -6.00 [-12.98, 0.98] | |
| Wu 2018 | 8 | 10 | 21 | 8 | 10.25 | 23 | 21.9% | 0.00 [-5.99, 5.99] | |
| Zheng 2018 | 15 | 5.5625 | 20 | 20 | 8.4375 | 22 | 42.6% | -5.00 [-9.29, -0.71] | |
| Total (95% CI) | 99 | 103 | 100.0% | -3.10 [-5.90, -0.30] | |||||
| Heterogeneity: Chi? = 3.36, df = 3 (P = 0.34); 1} = 11% Test for overall effect: Z = 2.17 (P = 0.03) | |||||||||
| -10 -5 0 5 10 | |||||||||
| Favours [LS] Favours [OS] | |||||||||
(b)
10
4
100
| (c) Study or Subgroup | Mean | LS SD | Total | Mean | OS SD | Total | Weight | Mean Difference IV, Random, 95% CI | Mean Difference IV, Random, 95% CI | |
|---|---|---|---|---|---|---|---|---|---|---|
| Brix 2010 | 6.2 | 1.75 | 35 | 7 | 1.875 | 35 | 10.8% | -0.80 | [-1.65, 0.05] | |
| Cooper 2013 | 8 | 3.5 | 46 | 12.05 | 5.68 | 256 | 10.0% | -4.05 | [-5.28, -2.82] | |
| Donatini 2014 | 5.5 | 1.65 | 13 | 6.8 | 1.125 | 21 | 10.5% | -1.30 | -2.32, -0.28] | |
| Fossa 2013 | 8 | 2.95 | 17 | 13 | 4.5 | 15 | 6.8% | -5.00 | [-7.67, -2.33] | |
| Huynh 2016 | 8 | 5.8 | 137 | 12.7 | 7.1 | 286 | 9.9% | -4.70 | [-5.97, -3.43] | |
| Kastelan 2020 | 7.5 | 2.1 | 23 | 12 | 4.5 | 23 | 8.2% | -4.50 | [-6.53, -2.47] | |
| Lombardi 2012 | 7.73 | 3.43 | 30 | 9.04 | 4.63 | 126 | 9.5% | -1.31 | [-2.78, 0.16] | |
| Mir 2013 | 7 | 4.375 | 18 | 13 | 4 | 26 | 7.1% | -6.00 | [-8.54, -3.46] | |
| Porpiglia 2010 | 9 | 3.25 | 18 | 10.5 | 3.5 | 25 | 8.2% | -1.50 | [-3.53, 0.53] | |
| Wu 2018 | 5.8 | 1.94 | 21 | 6.87 | 2.09 | 23 | 10.1% | -1.07 | [-2.26, 0.12] | |
| Zheng 2018 | 6.3 | 2.2 | 20 | 10.1 | 3.6 | 22 | 8.8% | -3.80 | [-5.59, -2.01] | |
| Total (95% CI) | 378 | 858 | 100.0% | -2.92 | [-4.02, -1.83] | |||||
| Heterogeneity: Tau2 = 2.70; Chi2 = 62.87, df = 10 (P < 0.00001); | I2 = 84% | -10 | -5 0 5 | |||||||
| Test for overall effect: Z = 5.25 (P < 0.00001) | Favours [LS] Favours [OS] | |||||||||
| (d) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mean | LS | Mean | OS | Weight | Mean Difference | Mean Difference IV, Random, 95% CI | ||||
| Study or Subgroup | SD | Total | SD | Total | IV, | Random, 95% CI | ||||
| Brix 2010 | 6 | 1.75 | 35 | 4 | 1.75 | 35 | 15.3% | 2.00 | [1.18, 2.82] | |
| Donatini 2014 | 4.3 | 1 | 13 | 4.32 | 1 | 21 | 16.5% | -0.02 | [-0.71, 0.67] | |
| Kastelan 2020 | 6 | 1.5 | 23 | 6 | 1.5 | 23 | 14.9% | 0.00 | [-0.87, 0.87] | |
| Mir 2013 | 7 | 0.5 | 18 | 7 | 0.3125 | 26 | 19.6% | 0.00 | [-0.26, 0.26] | |
| Porpiglia 2010 | 6 | 1.25 | 18 | 6 | 1.25 | 25 | 15.9% | 0.00 | [-0.76, 0.76] | |
| Wu 2018 | 4 | 1 | 21 | 5 | 0.75 | 23 | 17.9% | -1.00 | [-1.53, -0.47] | |
| Total (95% CI) | 128 | 153 | 100.0% | 0.12 | [-0.52, 0.77] | |||||
| Heterogeneity: Tau2 = | 0.54; | Chi2 = | 36.66, | df = 5 | (P < | 0.00001); I2 | = 86% | -2 0 2 | ||
| Test for overall effect: Z = | 0.38 (P | = 0.71) | 4 | |||||||
| Favours [LS] Favours [OS] | ||||||||||
| (e) | LS | OS | Odds Ratio | Odds Ratio | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, Fixed, 95% CI | M-H, | Fixed, | 95% CI | |
| Brix 2010 | 8 | 35 | 7 | 35 | 11.4% | 1.19 [0.38, 3.72] | ||||
| Cooper 2013 | 16 | 46 | 52 | 256 | 21.7% | 2.09 [1.06, 4.13] | ||||
| Donatini 2014 | 8 | 13 | 15 | 21 | 9.3% | 0.64 [0.15, 2.77] | ||||
| Kastelan 2020 | 11 | 23 | 16 | 23 | 17.5% | 0.40 [0.12, 1.34] | ||||
| Lombardi 2012 | 5 | 30 | 35 | 126 | 23.6% | 0.52 [0.18, 1.47] | ||||
| Mir 2013 | 9 | 18 | 9 | 26 | 7.7% | 1.89 [0.55, 6.45] | ||||
| Porpiglia 2010 | 12 | 18 | 15 | 25 | 8.8% | 1.33 [0.38, 4.72] | ||||
| Total (95% CI) | 183 | 512 | 100.0% | 1.10 [0.75, 1.63] | ||||||
| Total events | 69 | 149 | ||||||||
| Heterogeneity: Chi2 = 9.50, df = 6 (P = | 0.15); I2 | = 37% | 0.01 | 0.1 | 1 10 | |||||
| Test for overall effect: Z = 0.50 | (P = 0.62) | Favours | [LS] | Favours [OS] | ||||||
FIG. 1. Pooled estimate of baseline clinical characteristics between laparoscopic surgery and open surgery: a incidentaloma; b Ki67; c tumor size; d Weiss score; and e adjuvant mitotane. LS laparoscopic
comparable between the two groups. The pooled estimate of the Ki67 percentage score was evaluated among four stud- ies,24,25,33,34 with the LS group having a lower score than the OS group (MD - 3.10%, 95% CI - 5.90 to - 0.30; 22 = 11%). Lastly, adjuvant mitotane therapy was compared between the LS and OS groups, and no difference was observed between the two groups (OR 1.10, 95% CI 0.75-1.63; 12 = 37%). The baseline clinical characteristics of the included studies are shown in Fig. 1 and Table 2.
Survival Rates
Overall survival rates of the LS and OS groups were extracted and analyzed from the published Kaplan-Meier curve using Guyot’s method.20 The pooled estimate of 1-year (OR 0.74, 95% CI 0.54-1.01; 12 = 40%) and 2-year (OR 0.82, 95% CI 0.63-1.07; 12 = 0%) overall survival rates was comparable between the two groups. Conversely, the OS cohort had more favorable overall survival rates at 3 years than the LS cohort (OR 0.56, 95% CI 0.44-0.72; I 2 = 0%). Additionally, overall survival rates were evalu- ated at 5 years (OR 0.84, 95% CI 0.66-1.07; 12 = 17%) and 10 years (OR 0.83, 95% CI 0.43-1.58; 12 = 0%), and no difference was observed between the LS and OS groups. Similarly, the HR of the overall survival rates was com- parable between the two groups (HR 1.08, 95% CI 0.92-1.27; 12 = 33%). The overall survival rates of the included studies are comprehensively illustrated in Fig. 2.
In addition to overall survival rates, a subgroup analysis was performed on survival rates with a negative resection
surgery, OS open surgery, CI confidence interval, df degrees of freedom, SD standard deviation, M-H Mantel-Haenszel, IV inverse variance
margin between the LS and OS groups. The pooled esti- mate of 1-year (OR 0.83, 95% CI 0.27-2.50; 12 = 63%) and 2-year (OR 0.77, 95% CI 0.55-1.09; 12 = 34%) survival rates with a negative resection margin was comparable between the LS and OS groups. Similarly, no difference in survival rates with a negative resection margin was observed at 3 years (OR 0.86, 95% CI 0.62-1.19; 12 = 13%) and 5 years (OR 0.91, 95% CI 0.67-1.24; I 2 = 0%) between the two groups. Furthermore, the HR of survival rates with a negative resection margin was similar between the LS and OS groups (HR 0.95, 95% CI 0.76-1.18; 12 = 19%). The survival rates of the included studies with a negative resection margin are comprehen- sively illustrated in Fig. 3.
Recurrence-Free Rates
Similar to survival rates, recurrence-free rates between the LS and OS groups were extracted and analyzed from the published Kaplan-Meier curve using Guyot’s method.2º The pooled estimate of 1-year (OR 0.76, 95% CI 0.52-1.10; 7 = 0%) and 2-year (OR 0.88, 95% CI 0.61-1.25; 12 = 43%) recurrence-free rates was similar between the two groups. Contrarily, the recurrence-free rates at 3 years were more favorable for OS than LS (OR 0.60, 95% CI 0.42-0.85; 12 = 38%). Recurrence-free rates were analyzed at 5 years (OR 0.91, 95% CI 0.46-1.81; 12 = 55%) and 10 years (OR 1.08, 95% CI 0.32-3.66; I 2 = 72%), and no difference was observed between the two groups. Likewise, the HR of the recurrence-free rate
| Study | No. of patients | Tumor size, cm [mean ± SD] | Localization: right side [n (%)] | Adrenal incidentalomas [n (%)] | Surgical approach: transperitoneal [n (%)] | Adjuvant therapy: mitotane [n (%)] | Weiss score (mean ± SD) | Ki-67 score (mean ± SD) | AJCC stage 1 [n (%)] | AJCC stage 2 [n (%)] | AJCC stage 3 [n (%)] |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Brix et al.24 | LS: 35 | LS: 6.2 ± 1.8 | LS: 26 (74) | LS: 8 (23) | LS: 33 (94) | LS: 8 (23) | LS: 6 ± 2 | LS: 10 ± 12 | LS: 12 (34) | LS: 19 (54) | LS: 4 (11) |
| OS: 35 | OS: 7.0 ± 1.9 | OS: 17 (49) | OS: 9 (26) | OS: 19 (54) | OS: 7 (20) | OS: 4 ± 2 | OS: 10 ± 15 | OS: 9 (26) | OS: 22 (63) | OS: 4 (11) | |
| Cooper et al.28 | LS: 46 | LS: 8.0 ± 3.5 | LS: 19 (41) | LS: NR | LS: NR | LS: 16 (35) | LS: NR | LS: NR | LS: NR | LS: NR | LS: NR |
| OS: 256 | OS: 12.1 ± 5.7 | OS: 115 (45) | OS: NR | OS: NR | OS: 52 (20) | OS: NR | OS: NR | OS: NR | OS: NR | OS: NR | |
| Donatini et al.29 | LS: 13 | LS: 5.5 ± 1.7 | LS: NR | LS: NR | LS: NR | LS: 8 (62) | LS: 4.3 ± 1.0 | LS: NR | LS: NRº | LS: NRº | LS: 0 |
| OS: 21 | OS: 6.8 ± 1.1 | OS: NR | OS: NR | OS: NR | OS: 15 (71) | OS: 4.3 ± 1.0 | OS: NR | OS: NRº | OS: NRº | OS: 0 | |
| Fossa et al.30 | LS: 17 | LS: 8 ± 3 | LS: 10 (59) | LS: 4 (24) | LS: 17 (100) | LS: NR | LS: NR | LS: NR | LS: 1 (6) | LS: 12 (71) | LS: 4 (24) |
| OS: 15 | OS: 13 ± 5 | OS: 7 (47) | OS: 1 (7) | OS: 10 (67) | OS: NR | OS: NR | OS: NR | OS: 0 | OS: 7 (47) | OS: 8 (53) | |
| Kastelan et al.25 | LS: 23 | LS: 7.5 ± 2.1 | LS: NR | LS: NR | LS: 23 (100) | LS: 11 (48) | LS: 6 ± 1.5 | LS: 10 ± 6 | LS: 6 (26) | LS: 14 (61) | LS: 3 (13) |
| OS: 23 | OS: 12 ± 5 | OS: NR | OS: NR | OS: 23 (100) | OS: 16 (70) | OS: 6 ± 1.5 | OS: 16 ± 16 | OS: 0 | OS: 14 (61) | OS: 9 (39) | |
| Lombardi et al.27 | LS: 30 | LS: 7.73 ± 3.43 | LS: 23 (77) | LS: 25 (83) | LS: 29 (97) | LS: 5 (17) | LS: NR | LS: NR | LS: 10 (33) | LS: 20 (67) | LS: 0 |
| OS: 126 | OS: 9.04 ± 4.63 | OS: 64 (51) | OS: 59 (47) | OS: NR | OS: 35 (28) | OS: NR | OS: NR | OS: 24 (19) | OS: 102 (81) | OS: 0 | |
| Mir et al.31 | LS: 18 | LS: 7.0 ± 4.4 | LS: 17 (94) | LS: NR | LS: NR | LS: 9 (50) | LS: 7.0 ± 0.5 | LS: NR | LS: NRd | LS: NRd | LS: NRd |
| OS: 26 | OS: 13.0 ± 4.0 | OS: 11 (42) | OS: NR | OS: NR | OS: 9 (35) | OS: 7.0 ± 0.3 | OS: NR | OS: NRd | OS: NRd | OS: NRd | |
| Porpiglia et al. 32 | LS: 18 | LS: 9.0 ± 3.3 | LS: NR | LS: 6 (33) | LS: 18 (100) | LS: 12 (67) | LS: 6 ± 1 | LS: NR | LS: 3 (17) | LS: 15 (83) | LS: 0 |
| OS: 25 | OS: 10.5 ± 3.5 | OS: NR | OS: 7 (28) | OS: 25 (100) | OS: 15 (60) | OS: 6 ± 1 | OS: NR | OS: 3 (12) | OS: 22 (88) | OS: 0 | |
| Tseng et al. 26 | LS: 231 | LS: NRª | LS: 105 (45) | LS: NR | LS: NR | LS: NR | LS: NR | LS: NR | LS: NR | LS: NR | LS: NR |
| OS: 573 | OS: NRª | OS: 273 (48) | OS: NR | OS: NR | OS: NR | OS: NR | OS: NR | OS: NR | OS: NR | OS: NR | |
| Wu et al. 33 | LS: 21 | LS: 5.80 ± 1.94 | LS: 9 (43) | LS: NR | LS: 11 (52) | LS: NRb | LS: 4 ± 1 | LS: 8 ± 10 | LS: 6 (29) | LS: 15 (71) | LS: 0 |
| OS: 23 | OS: 6.87 ± 2.09 | OS: 6 (26) | OS: NR | OS: NR | OS: NRb | OS: 5 ± 1 | OS: 8 ± 10 | OS: 3 (13) | OS: 20 (87) | OS: 0 | |
| Zheng et al. 34 | LS: 20 | LS: 6.3 ± 2.2 | LS: NR | LS: NR | LS: NR | LS: NR | LS: NR | LS: 15 ± 6 | LS: 5 (25) | LS: 11 (55) | LS: 4 (20) |
| OS: 22 | OS: 48.5 ± 3.6 | OS: NR | OS: NR | OS: NR | OS: NR | OS: NR | OS: 20 ± 8 | OS: 1 (5) | OS: 12 (55) | OS: 9 (41) |
AJCC American Joint Committee of Cancer, LS laparoscopic surgery, N number of patients, NR not reported, OS open surgery, SD standard deviation
ªLS: < 4 cm (n = 25, 10.8%), 4-6 cm (n = 58, 25.1%), > 6 cm (n = 148, 64.1%); OS: < 4 cm (n = 47, 8.2%), 4-6 cm (n = 41, 7.2%), > 6 cm (n = 485, 84.6%) “Number of adjuvant therapies for LS (n = 2, 9.5%) and OS (n = 4, 17.4%)
“LS: stage I/II (n = 13, 100%); OS: stage I/II (n = 21, 100%)
dLS: stage I/II (n = 15, 82%), stage III/IV (n = 3, 11%); OS: stage I/II (n = 9, 35%), stage III/IV (n = 17, 62%)
| (a) | ||||||||
|---|---|---|---|---|---|---|---|---|
| LS | OS | Odds Ratio | Odds | Ratio | ||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, Fixed, 95% CI | M-H, Fixed, | 95% CI |
| Cooper 2013 | 40 | 46 | 223 | 256 | 10.5% | 0.99 [0.39, 2.51] | ||
| Donatini 2014 | 13 | 13 | 19 | 21 | 0.6% | 3.46 [0.15, 77.97] | ||
| Fossa 2013 | 14 | 17 | 15 | 15 | 3.8% | 0.13 [0.01, 2.82] | ||
| Kastelan 2020 | 22 | 23 | 21 | 23 | 1.1% | 2.10 [0.18, 24.87] | ||
| Lombardi 2012 | 20 | 30 | 115 | 126 | 17.5% | 0.19 [0.07, 0.51] | ||
| Mir 2013 | 14 | 18 | 17 | 26 | 3.7% | 1.85 [0.47, 7.32] | ||
| Porpiglia 2010 | 18 | 18 | 23 | 25 | 0.6% | 3.94 [0.18, 87.10] | ||
| Tseng 2022 | 170 | 213 | 443 | 525 | 61.1% | 0.73 [0.49, 1.10] | ||
| Wu 2018 | 20 | 21 | 22 | 23 | 1.2% | 0.91 [0.05, 15.52] | ||
| Total (95% CI) | 399 | 1040 | 100.0% | 0.74 [0.54, 1.01] | ||||
| Total events | 331 | 898 | ||||||
| Heterogeneity: Chi2 = 13.38, df | = 8 (P | = 0.10); | I2 = 40% | |||||
| Test for overall effect: | Z = 1.88 | (P = 0.06) | 0.01 | 0.1 | 1 10 | |||
| Favours [OS] | Favours [LS] | |||||||
100
| (c) | LS | OS | Weight | Odds Ratio | Odds | Ratio | (d) | |||
|---|---|---|---|---|---|---|---|---|---|---|
| Study or Subgroup | Events | Total | Events | Total | M-H, | Fixed, 95% CI | M-H, Fixed, | 95% CI | ||
| Cooper 2013 | 25 | 46 | 190 | 256 | 17.2% | 0.41 [0.22, 0.79] | ||||
| Donatini 2014 | 11 | 13 | 18 | 21 | 1.4% | 0.92 [0.13, 6.38] | ||||
| Fossa 2013 | 12 | 17 | 11 | 15 | 2.2% | 0.87 [0.19, 4.11] | ||||
| Kastelan 2020 | 21 | 23 | 21 | 23 | 1.2% | 1.00 [0.13, 7.78] | ||||
| Lombardi 2012 | 17 | 30 | 96 | 126 | 10.4% | 0.41 [0.18, 0.94] | ||||
| Mir 2013 | 9 | 18 | 14 | 26 | 3.7% | 0.86 [0.26, 2.86] | ||||
| Porpiglia 2010 | 18 | 18 | 21 | 25 | 0.3% | 7.74 [0.39, 153.56] | ||||
| Tseng 2022 | 124 | 213 | 374 | 525 | 58.7% | 0.56 | [0.40, 0.78] | |||
| Wu 2018 | 12 | 21 | 18 | 23 | 4.8% | 0.37 [0.10, 1.38] | ||||
| Total (95% CI) | 399 | 1040 | 100.0% | 0.56 | [0.44, 0.72] | |||||
| Total events | 249 | 763 | ||||||||
| Heterogeneity: Chi2 = 6.12, df | = 8 (P | = 0.63); | I= 0% | 1 10 | 100 | |||||
| Test for overall effect: | Z = 4.45 | (P | <0.00001) | 0.01 | 0.1 | |||||
| Favours [OS] | Favours [LS] | |||||||||
| (e) | LS | OS | Odds Ratio | Odds | Ratio | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, | Fixed, 95% CI | M-H, Fixed, | 95% CI | ||
| Donatini 2014 | 9 | 13 | 18 | 21 | 20.9% | 0.38 | [0.07, 2.05] | ||||
| Kastelan 2020 | 19 | 23 | 17 | 23 | 14.6% | 1.68 | [0.40, 6.97] | ||||
| Lombardi 2012 | 6 | 30 | 27 | 126 | 41.0% | 0.92 | [0.34, 2.47] | Porpiglia 2010 | |||
| Mir 2013 | 3 | 18 | 7 | 26 | 23.5% | 0.54 | [0.12, 2.46] | ||||
| Total (95% CI) | 84 | 196 | 100.0% | 0.83 | [0.43, 1.58] | ||||||
| Total events | 37 | 69 | Total (95% CI) | ||||||||
| Heterogeneity: Chi2 = | 2.12, df | = 3 (P = | 0.55); I' | = 0% | 5 | Heterogeneity: Chi2 = 11.93, df = 8 (P = 0.15); I' = 33% | |||||
| Test for overall effect: | Z = 0.57 | (P = | 0.57) | 0.05 | 0.2 1 | 20 | Test for overall effect: Z = 0.98 (P = 0.33) | ||||
| Favours [OS] | Favours [LS] | ||||||||||
FIG. 2. Pooled estimate of the overall survival rate between laparoscopic surgery and open surgery at a 1 year; b 2 years; c 3 years; d 5 years; e 10 years; and f hazard ratio. LS laparoscopic
| (a) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| LS | OS | Odds Ratio | Odds Ratio M-H, Random, 95% CI | |||||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, | Random, 95% CI | |||
| Donatini 2014 | 13 | 13 | 19 | 21 | 9.6% | 3.46 [0.15, 77.97] | ||||
| Kastelan 2020 | 22 | 23 | 21 | 23 | 13.3% | 2.10 | [0.18, 24.87] | |||
| Lombardi 2012 | 20 | 30 | 115 | 126 | 30.5% | 0.19 | [0.07, 0.51] | |||
| Porpiglia 2010 | 18 | 18 | 23 | 25 | 9.7% | 3.94 | [0.18, 87.10] | |||
| Tseng 2022 | 142 | 165 | 381 | 437 | 36.9% | 0.91 [0.54, 1.53] | ||||
| Total (95% CI) | 249 | 632 | 100.0% | 0.83 | [0.27, 2.50] | |||||
| Total events | 215 | 559 | ||||||||
| Heterogeneity: Tau2 | = 0.79; Chi2 | = | 10.72, df = | 4 (P = | 0.03); I' = | 63% | ||||
| Test for overall effect: Z = 0.34 (P = | 0.74) | 0.01 | 0.1 1 | 10 | ||||||
| Favours [OS] Favours [LS] | ||||||||||
| (c) | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| LS | OS | Odds Ratio | Odds | Ratio | |||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, Fixed, 95% CI | M-H, Fixed, | 95% CI | |
| Donatini 2014 | 11 | 13 | 18 | 21 | 2.7% | 0.92 [0.13, 6.38] | |||
| Kastelan 2020 | 21 | 23 | 21 | 23 | 2.3% | 1.00 [0.13, 7.78] | |||
| Lombardi 2012 | 17 | 30 | 90 | 126 | 19.3% | 0.52 [0.23, 1.19] | |||
| Porpiglia 2010 | 18 | 18 | 19 | 25 | 0.6% | 12.33 [0.65, 234.69] | |||
| Tseng 2022 | 105 | 165 | 293 | 437 | 75.1% | 0.86 [0.59, 1.25] | |||
| Total (95% CI) | 249 | 632 | 100.0% | 0.86 [0.62, 1.19] | |||||
| Total events | 172 | 441 | |||||||
| Heterogeneity: Chi2 = 4.59, df | = 4 (P = | 0.33); Iª | = 13% | 0.01 | 0.1 1 | 10 | |||
| Test for overall effect: Z = 0.88 | (P = | 0.38) | Favours [OS] | Favours [LS] | |||||
| (e) | ||||||||
|---|---|---|---|---|---|---|---|---|
| Hazard Ratio | Hazard | Ratio | ||||||
| Study or Subgroup | log[Hazard Ratio] | SE | Weight | IV, Fixed, 95% CI | IV, Fixed, | 95% CI | ||
| Donatini 2014 | 0.7311 | 0.7644 | 2.2% | 2.08 [0.46, 9.29] | ||||
| Kastelan 2020 | -0.3913 | 0.6457 | 3.1% | 0.68 [0.19, 2.40] | ||||
| Lombardi 2012 | -0.4872 | 0.2657 | 18.4% | 0.61 [0.36, 1.03] | ||||
| Porpiglia 2010 | -0.3606 | 0.6273 | 3.3% | 0.70 [0.20, 2.38] | ||||
| Tseng 2022 | 0.0582 | 0.1333 | 73.0% | 1.06 [0.82, 1.38] | ||||
| Total (95% CI) | 100.0% | 0.95 [0.76, 1.18] | ||||||
| Heterogeneity: Chi2 = 4.93, df = 4 (P = 0.29); I2 = 19% | 10 | |||||||
| Test for overall effect: Z = 0.48 (P = 0.63) | 0.01 | 0.1 Favours [LS] | 1 Favours [OS] | |||||
FIG. 3. Pooled estimate of the overall survival rate with negative resection margin between laparoscopic surgery and open surgery at a 1 year; b 2 years; c 3 years; d 5 years; and e hazard ratio. LS
was comparable between the LS and OS cohorts (HR 1.22, 95% CI 0.81-1.83; 12 = 69%). The recurrence-free rates of the included studies are comprehensively illustrated in Fig. 4.
| (b) | ||||||||
|---|---|---|---|---|---|---|---|---|
| Study or Subgroup | LS | OS | Weight | Odds Ratio M-H, Fixed, 95% CI | Odds Ratio M-H, Fixed, 95% CI | |||
| Events | Total | Events | Total | |||||
| Cooper 2013 | 32 | 46 | 190 | 256 | 14.6% | 0.79 [0.40, 1.58] | ||
| Donatini 2014 | 13 | 13 | 18 | 21 | 0.4% | 5.11 [0.24, 107.33] | ||
| Fossa 2013 | 12 | 17 | 11 | 15 | 2.8% | 0.87 [0.19, 4.11] | ||
| Kastelan 2020 | 21 | 23 | 21 | 23 | 1.5% | 1.00 [0.13, 7.78] | ||
| Lombardi 2012 | 17 | 30 | 96 | 126 | 13.2% | 0.41 [0.18, 0.94] | ||
| Mir 2013 | 12 | 18 | 14 | 26 | 3.2% | 1.71 [0.49, 5.97] | ||
| Porpiglia 2010 | 18 | 18 | 21 | 25 | 0.4% | 7.74 [0.39, 153.56] | ||
| Tseng 2022 | 140 | 213 | 374 | 525 | 61.2% | 0.77 [0.55, 1.09] | ||
| Wu 2018 | 17 | 21 | 18 | 23 | 2.7% | 1.18 [0.27, 5.15] | ||
| Total (95% CI) | 399 | 1040 | 100.0% | 0.82 [0.63, 1.07] | ||||
| Total events | 282 | 763 | ||||||
| Heterogeneity: Chi2 = 8.00, df = 8 (P = 0.43); I' | = 0% | |||||||
| Test for overall effect: Z = 1.47 (P = 0.14) | 0.01 | 0.1 1 10 Favours [LS] Favours [OS] | ||||||
100
| LS | OS | Weight | Odds Ratio | Odds Ratio | ||||
|---|---|---|---|---|---|---|---|---|
| Study or Subgroup | Events | Total | Events | Total | M-H, Fixed, 95% CI | M-H, Fixed, | 95% CI | |
| Cooper 2013 | 16 | 46 | 123 | 256 | 17.2% | 0.58 [0.30, 1.11] | ||
| Donatini 2014 | 11 | 13 | 18 | 21 | 1.5% | 0.92 [0.13, 6.38] | ||
| Fossa 2013 | 10 | 17 | 3 | 15 | 0.9% | 5.71 [1.16, 28.07] | ||
| Kastelan 2020 | 19 | 23 | 21 | 23 | 2.6% | 0.45 [0.07, 2.76] | ||
| Lombardi 2012 | 13 | 30 | 60 | 126 | 9.2% | 0.84 [0.38, 1.88] | ||
| Mir 2013 | 4 | 18 | 9 | 26 | 4.0% | 0.54 [0.14, 2.13] | ||
| Porpiglia 2010 | 14 | 18 | 18 | 25 | 2.4% | 1.36 [0.33, 5.59] | ||
| Tseng 2022 | 97 | 213 | 267 | 525 | 59.2% | 0.81 [0.59, 1.11] | ||
| Wu 2018 | 12 | 21 | 10 | 23 | 2.9% | 1.73 [0.53, 5.72] | ||
| Total (95% CI) | 399 | 1040 | 100.0% | 0.84 [0.66, 1.07] | ||||
| Total events | 196 | 529 | ||||||
| Heterogeneity: Chi2 = 9.61, df = 8 (P = 0.29); I2 | = 17% | 1 10 | ||||||
| Test for overall effect: Z = 1.42 (P = 0.16) | 0.01 | 0.1 | ||||||
| Favours [LS] | Favours [OS] | |||||||
100
| (f) | Hazard Ratio | Hazard Ratio | ||||
|---|---|---|---|---|---|---|
| Study or Subgroup | log[Hazard Ratio] | SE | Weight | IV, Fixed, 95% CI | IV, Fixed, | 95% CI |
| Cooper 2013 | 0.2946 | 0.1928 | 18.3% | 1.34 [0.92, 1.96] | ||
| Donatini 2014 | 0.7311 | 0.7644 | 1.2% | 2.08 [0.46, 9.29] | ||
| Fossa 2013 | -0.7102 | 0.4804 | 2.9% | 0.49 [0.19, 1.26] | ||
| Kastelan 2020 | -0.3913 | 0.6457 | 1.6% | 0.68 [0.19, 2.40] | ||
| Lombardi 2012 | -0.4872 | 0.2657 | 9.6% | 0.61 [0.36, 1.03] | ||
| Mir 2013 | 0.4001 | 0.3346 | 6.1% | 1.49 [0.77, 2.87] | ||
| -0.3606 | 0.6273 | 1.7% | 0.70 [0.20, 2.38] | |||
| Tseng 2022 | 0.1488 | 0.1115 | 54.6% | 1.16 [0.93, 1.44] | ||
| Wu 2018 | -0.1721 | 0.41 | 4.0% | 0.84 [0.38, 1.88] | ||
| 100.0% | 1.08 [0.92, 1.27] | |||||
| 0.01 | 0.1 | 1 10 | ||||
| Favours [LS] | Favours [OS] | |||||
100
surgery, OS open surgery, CI confidence interval, df degrees of freedom, M-H Mantel-Haenszel, IV inverse variance, SE standard error
100
| (b) | |||||||
|---|---|---|---|---|---|---|---|
| Experimental | Control | Odds Ratio | Odds Ratio | ||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, Fixed, 95% CI | M-H, Fixed, 95% CI |
| Donatini 2014 | 13 | 13 | 18 | 21 | 0.7% | 5.11 [0.24, 107.33] | |
| Kastelan 2020 | 21 | 23 | 21 | 23 | 2.5% | 1.00 [0.13, 7.78] | |
| Lombardi 2012 | 17 | 30 | 96 | 126 | 22.3% | 0.41 [0.18, 0.94] | |
| Porpiglia 2010 | 18 | 18 | 21 | 25 | 0.7% | 7.74 [0.39, 153.56] | |
| Tseng 2022 | 117 | 165 | 332 | 437 | 73.8% | 0.77 [0.52, 1.15] | |
| Total (95% CI) | 249 | 632 | 100.0% | 0.77 [0.55, 1.09] | |||
| Total events | 186 | 488 | |||||
| Heterogeneity: Chi2 = 6.09, df = 4 (P = | 0.19); I2 | = 34% | |||||
| Test for overall effect: Z = 1.46 (P = 0.14) | 0.01 | 0.1 1 10 Favours [OS] Favours [LS] | |||||
| (d) | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| LS | OS | Odds | Ratio | Odds | Ratio | ||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, | Fixed, 95% CI | M-H, Fixed, | 95% CI |
| Donatini 2014 | 11 | 13 | 18 | 21 | 2.5% | 0.92 | [0.13, 6.38] | ||
| Kastelan 2020 | 19 | 23 | 21 | 23 | 4.3% | 0.45 | [0.07, 2.76] | ||
| Lombardi 2012 | 13 | 30 | 60 | 126 | 15.5% | 0.84 | [0.38, 1.88] | ||
| Porpiglia 2010 | 14 | 18 | 18 | 25 | 4.0% | 1.36 | [0.33, 5.59] | ||
| Tseng 2022 | 87 | 165 | 239 | 437 | 73.6% | 0.92 | [0.65, 1.32] | ||
| Total (95% CI) | 249 | 632 | 100.0% | 0.91 | [0.67, 1.24] | ||||
| Total events | 144 | 356 | |||||||
| Heterogeneity: Chi2 = 0.93, df | = 4 (P | = 0.92); I2 | = 0% | ||||||
| Test for overall effect: Z = 0.61 | (P = | 0.54) | 0.01 | 0.1 Favours [OS] | 1 10 Favours [LS] | ||||
100
laparoscopic surgery, OS open surgery, CI confidence interval, df degrees of freedom, M-H Mantel-Haenszel, SE standard error
Perioperative Outcomes
In addition to primary outcomes, perioperative out- comes were evaluated between the LS and OS groups. Five comparative studies 27,30,31,33,34 reported the operative time,
100
100
100
| (a) | LS | OS | Odds Ratio | (b) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Odds Ratio | ||||||||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, Fixed, 95% CI | M-H, Fixed, | 95% CI | ||
| Brix 2010 | 23 | 35 | 24 | 35 | 12.8% | 0.88 [0.32, 2.38] | ||||
| Cooper 2013 | 17 | 46 | 127 | 256 | 38.0% | 0.60 [0.31, 1.14] | . | |||
| Donatini 2014 | 13 | 13 | 20 | 21 | 0.9% | 1.98 [0.07, 52.16] | ||||
| Kastelan 2020 | 22 | 23 | 22 | 23 | 1.5% | 1.00 [0.06, 17.02] | ||||
| Lombardi 2012 | 22 | 30 | 102 | 126 | 16.3% | 0.65 [0.26, 1.63] | ||||
| Mir 2013 | 10 | 18 | 18 | 26 | 10.2% | 0.56 [0.16, 1.94] | ||||
| Porpiglia 2010 | 13 | 18 | 15 | 25 | 5.4% | 1.73 [0.47, 6.39] | ||||
| Wu 2018 | 14 | 21 | 12 | 23 | 5.9% | 1.83 [0.54, 6.22] | ||||
| Zheng 2018 | 14 | 20 | 20 | 22 | 8.9% | 0.23 [0.04, 1.33] | ||||
| Total (95% CI) | 224 | 557 | 100.0% | 0.76 [0.52, 1.10] | ||||||
| Total events | 148 | 360 | ||||||||
| Heterogeneity: Chi2 = 6.65, df = 8 (P = 0.58); I | = 0% | 0.01 | 0.1 | 1 10 | 100 | |||||
| Test for overall effect: Z = 1.47 (P = 0.14) | ||||||||||
| Favours [OS] Favours [LS] | ||||||||||
| (c) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| LS | OS | Odds | Ratio | Odds | Ratio | |||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, | Fixed, 95% CI | M-H, Fixed, | 95% CI | |
| Brix 2010 | 14 | 35 | 18 | 35 | 13.5% | 0.63 | [0.24, 1.62] | |||
| Cooper 2013 | 9 | 46 | 84 | 256 | 26.0% | 0.50 | [0.23, 1.08] | |||
| Donatini 2014 | 10 | 13 | 18 | 21 | 4.0% | 0.56 | [0.09, 3.29] | |||
| Kastelan 2020 | 20 | 23 | 20 | 23 | 3.3% | 1.00 | [0.18, 5.56] | |||
| Lombardi 2012 | 19 | 30 | 78 | 126 | 13.9% | 1.06 | [0.47, 2.43] | |||
| Mir 2013 | 5 | 18 | 14 | 26 | 10.5% | 0.33 | [0.09, 1.19] | |||
| Porpiglia 2010 | 8 | 18 | 11 | 25 | 6.5% | 1.02 | [0.30, 3.45] | |||
| Wu 2018 | 8 | 21 | 8 | 23 | 6.0% | 1.15 | [0.34, 3.95] | |||
| Zheng 2018 | 2 | 20 | 15 | 22 | 16.2% | 0.05 | [0.01, 0.29] | |||
| Total (95% CI) | 224 | 557 | 100.0% | 0.60 | [0.42, 0.85] | |||||
| Total events | 95 | 266 | ||||||||
| Heterogeneity: Chi2 = 12.92, df | = 8 (P | = 0.11); I2 = 38% | 0.01 | 0.1 | 1 10 | |||||
| Test for overall effect: Z = 2.84 | (P = | 0.005) | ||||||||
| Favours [OS] | Favours [LS] | |||||||||
100
| (e) Study or Subgroup | LS | OS | Odds Ratio | Odds Ratio M-H, Random, 95% CI | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Events | Total | Events | Total | Weight | M-I | Random, | 95% CI | |||
| Donatini 2014 | 7 | 13 | 14 | 21 | 21.2% | 0.58 [0.14, 2.41] | ||||
| Kastelan 2020 | 19 | 23 | 17 | 23 | 21.1% | 1.68 | [0.40, 6.97] | |||
| Lombardi 2012 | 10 | 30 | 10 | 126 | 24.8% | 5.80 | [2.14, 15.71] | |||
| Mir 2013 | 5 | 18 | 10 | 26 | 22.2% | 0.62 | [0.17, 2.26] | |||
| Zheng 2018 | 0 | 20 | 4 | 22 | 10.7% | 0.10 | [0.01, 1.99] | |||
| Total (95% CI) | 104 | 218 | 100.0% | 1.08 | [0.32, 3.66] | |||||
| Total events | 41 | 55 | ||||||||
| Heterogeneity: Tau2 = 1.31; Chi2 = | 14.09, df = | 4 (P = 0.007); I2 = | 72% | |||||||
| Test for overall effect: Z = 0.12 (P = 0.90) | 0.01 | 0.1 1 | 10 | |||||||
| Favours [OS] Favours [LS] | ||||||||||
100
and no difference was observed between the LS and OS cohorts (MD - 16.89 min, 95% CI - 50.88 to 17.09; 12 = 89%). Similarly, the intraoperative capsular rupture was analyzed in five studies,24,29,30,32,34 and the pooled estimate of capsular rupture was similar between the two groups (OR 1.26, 95% CI 0.55-2.86; 12 = 0%). Conversely, 10 studies24-32,34 reported the incidents of positive resec- tion margin, and the OS cohort had a lower rate of positive resection margin than the LS cohort (OR 1.52, 95% CI 1.10-2.10; 1 = 0%). Hospital stay was reported in six studies. 27,29-31,33,34 and the LS group had a shorter length of hospital stay than the OS group (MD - 2.49 days, 95% CI - 2.95 to - 2.04; I 2 = 45%). Additionally, six stud- ies
ies27,29-31,33,34 described postoperative complications. The LS cohort had 36 complications (30.3%), and the OS cohort reported 58 complications (24.9%). However, LS and OS had a comparable rate of postoperative complica- tions (OR 0.54, 95% CI 0.25-1.17; 12 = 0%). Lastly, 10 studies24-33 reported the conversion from laparoscopic to open, accounting for 15.7% (n = 71) of the laparoscopic group. The perioperative outcomes of the included studies are depicted in Fig. 5.
| LS | OS | Weight | ||||||
|---|---|---|---|---|---|---|---|---|
| Odds Ratio M-H, Fixed, 95% CI | Odds Ratio M-H, Fixed, 95% CI | |||||||
| Study or Subgroup | Events | Total | Events | Total | ||||
| Brix 2010 | 17 | 35 | 18 | 35 | 14.2% | 0.89 [0.35, 2.28] | ||
| Cooper 2013 | 10 | 46 | 84 | 256 | 30.8% | 0.57 [0.27, 1.20] | . | |
| Donatini 2014 | 12 | 13 | 18 | 21 | 1.6% | 2.00 [0.19, 21.57] | ||
| Kastelan 2020 | 22 | 23 | 20 | 23 | 1.3% | 3.30 [0.32, 34.35] | ||
| Lombardi 2012 | 22 | 30 | 78 | 126 | 12.3% | 1.69 [0.70, 4.10] | ||
| Mir 2013 | 7 | 18 | 14 | 26 | 10.8% | 0.55 [0.16, 1.85] | ||
| Porpiglia 2010 | 10 | 18 | 11 | 25 | 6.3% | 1.59 [0.47, 5.39] | ||
| Wu 2018 | 10 | 21 | 8 | 23 | 6.2% | 1.70 [0.51, 5.73] | ||
| Zheng 2018 | 5 | 20 | 15 | 22 | 16.5% | 0.16 [0.04, 0.60] | ||
| Total (95% CI) | 224 | 557 | 100.0% | 0.88 [0.61, 1.25] | ||||
| Total events | 115 | 266 | ||||||
| Heterogeneity: Chi2 = 14.03, df = 8 (P = 0.08); I2 = 43% Test for overall effect: Z = 0.73 (P = 0.47) | 0.02 | 0.1 1 10 | ||||||
| Favours [OS] Favours [LS] | ||||||||
50
(d)
| LS | OS | Odds Ratio | Odds | Ratio | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H I. | Random, 95% CI | M-H, Fixed, | 95% CI | ||
| Brix 2010 | 12 | 35 | 10 | 35 | 15.9% | 1.30 [0.47, 3.59] | |||||
| Cooper 2013 | 2 | 46 | 37 | 256 | 11.6% | 0.27 [0.06, 1.16] | |||||
| Donatini 2014 | 8 | 13 | 14 | 21 | 11.8% | 0.80 [0.19, 3.37] | |||||
| Kastelan 2020 | 19 | 23 | 17 | 23 | 11.9% | 1.68 [0.40, 6.97] | |||||
| Lombardi 2012 | 17 | 30 | 48 | 126 | 18.2% | 2.13 [0.95, 4.76] | |||||
| Mir 2013 | 5 | 18 | 10 | 26 | 13.0% | 0.62 [0.17, 2.26] | |||||
| Wu 2018 | 8 | 21 | 6 | 23 | 13.2% | 1.74 [0.48, 6.28] | |||||
| Zheng 2018 | 0 | 20 | 11 | 22 | 4.5% | 0.02 [0.00, 0.45] | |||||
| Total (95% CI) | 206 | 532 | 100.0% | 0.91 [0.46, 1.81] | |||||||
| Total events | 71 | 153 | |||||||||
| Heterogeneity: Tau2 = | 0.51; Chi2 | = | 15.57, df = | 7 (P = | 0.03); 1ª | = 55% | |||||
| Test for overall effect: Z = 0.27 (P = 0.79) | 0.01 | 0.1 | 1 10 | ||||||||
| Favours [OS] | Favours [LS] | ||||||||||
100
| (f) | log[Hazard Ratio] | SE | Hazard Ratio | Hazard Ratio | |||||
|---|---|---|---|---|---|---|---|---|---|
| Study or Subgroup | Weight | IV | Random, | 95% CI | IV, | Random, | 95% CI | ||
| Brix 2010 | -0.09153 | 0.28922 | 13.3% | 0.91 | [0.52, 1.61] | ||||
| Cooper 2013 | 0.3146 | 0.1654 | 16.1% | 1.37 | [0.99, 1.89] | ||||
| Donatini 2014 | 0.3762 | 0.6864 | 6.0% | 1.46 | [0.38, 5.59] | ||||
| Kastelan 2020 | -0.4611 | 0.6457 | 6.5% | 0.63 | [0.18, 2.24] | ||||
| Lombardi 2012 | -0.5756 | 0.245 | 14.3% | 0.56 | [0.35, 0.91] | ||||
| Mir 2013 | 0.3701 | 0.3754 | 11.2% | 1.45 | [0.69, 3.02] | ||||
| Porpiglia 2010 | 0.3764 | 0.3435 | 12.0% | 1.46 | [0.74, 2.86] | ||||
| Wu 2018 | -0.1109 | 0.3792 | 11.2% | 0.90 | [0.43, 1.88] | ||||
| Zheng 2018 | 1.841 | 0.4601 | 9.5% | 6.30 | [2.56, 15.53] | ||||
| Total (95% CI) | 100.0% | 1.22 | [0.81, 1.83] | ||||||
| Heterogeneity: Tau2 = | 0.24; Chi2 = 26.22, df | = 8 (P = | 0.0010); I' | = | 69% | 0.05 | 0.2 | 1 5 | |
| Test for overall effect: | Z = 0.96 (P = 0.34) | ||||||||
| Favours | [LS] | Favours [OS] | |||||||
20
surgery, CI confidence interval, df degrees of freedom, M-H Mantel- Haenszel, SE standard error, IV inverse variance
DISCUSSION
For ACC, curative therapy aims to ensure a safe nega- tive margin resection of the tumor mass with minimal complications.35,36 Such efforts are essential in minimizing the incidence of recurrence and ultimately improving the survival outcomes for patients. The primary aim of this meta-analysis was to investigate the safety and efficacy of LS and OS in managing ACC. Based on the 11 studies within this meta-analysis, the overall survival and recur- rence-free rates for 1, 2, 5, and 10 years were comparable between the two groups. In contrast, OS was more favor- able than LS at 3 years for the overall survival and recurrence-free rates. Additionally, no difference was observed in a subgroup analysis of survival rates with a negative resection margin at 1, 2, 3, and 5 years, as well as the HR between the two cohorts. Similarly, operative time, intraoperative capsular rupture, and postoperative compli- cations revealed no difference between LS and OS. Conversely, LS showed a shorter length of postoperative hospital stay, while OS had a lower rate of positive resection margins.
| (a) | LS | OS | Mean Difference | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean Difference | |||||||||||||
| Study or Subgroup | Mean | SD | Total | Mcan | 'D | Total | Weight | IV, | Random, 95% CI | IV, | Random, 95% CI | ||
| Fossa 2013 | 150 | 77.5 | 17 | 230 | 46 | 15 | 17.1% | -80.00 | [-123.58, -36.42] | ||||
| Lombardi 2012 | 135 | 65 | 30 | 129 | 54 | 126 | 21.1% | 6.00 | [-19.10, 31.10] | ||||
| Mir 2013 | 297.5 | 67.675 | 18 | 272.5 | 67.675 | 26 | 17.7% | 25.00 | [-15.67, 65.67] | ||||
| Wu 2018 | 125 | 35 | 21 | 117 | 56 | 23 | 20.6% | 8.00 | [-19.35, 35.35] | ||||
| Zheng 2018 | 130 | 13.375 | 20 | 175 | 13.625 | 22 | 23.5% | -45.00 | [-53.17, -36.83] | ||||
| Total (95% CI) | 106 | 212 | 100.0% | -16.89 | [-50.88, 17.09] | ||||||||
| Heterogeneity: Tau? = 1263.59; | Chi" = | 37.92, | df = 4 | (P < 0.00001); | I' = 89% | ||||||||
| Test for overall effect: Z = 0.97 (P = 0.33) | -100 | -50 | 50 0 | ||||||||||
100
Favours [LS] Favours [OS]
100
100
| (c) | ||||||||
|---|---|---|---|---|---|---|---|---|
| LS | OS | Weight | Odds Ratio | Odds Ratio | ||||
| Study or Subgroup | Events | Total | Events | Total | M-H, Fixed, 95% CI | M-H Fixed, | 95% CI | |
| Brix 2010 | 2 | 35 | 1 | 35 | 1.6% | 2.06 [0.18, 23.83] | ||
| Cooper 2013 | 13 | 46 | 41 | 256 | 15.7% | 2.07 [1.00, 4.26] | ||
| Donatini 2014 | 0 | 13 | 0 | 21 | Not estimable | |||
| Fossa 2013 | 5 | 17 | 3 | 15 | 3.9% | 1.67 [0.32, 8.59] | ||
| Kastelan 2020 | 0 | 23 | 0 | 23 | Not estimable | |||
| Lombardi 2012 | 0 | 30 | 0 | 126 | Not estimable | |||
| Mir 2013 | 7 | 18 | 10 | 26 | 8.7% | 1.02 [0.30, 3.50] | ||
| Porpiglia 2010 | 0 | 18 | 0 | 25 | Not estimable | |||
| Tseng 2022 | 48 | 231 | 88 | 573 | 70.0% | 1.45 [0.98, 2.14] | ||
| Zheng 2018 | 0 | 20 | 0 | 22 | Not estimable | |||
| Total (95% CI) | 451 | 1122 | 100.0% | 1.52 [1.10, 2.10] | ||||
| Total events | 75 | 143 | ||||||
| Heterogeneity: Chi2 = | 1.23, df | = 4 (P = | 0.87); 1ª | = 0% | ||||
| Test for overall effect: | Z = 2.56 | (P = | 0.01) | 0.01 | 0.1 | 1 10 | ||
| Favours [LS] | Favours [OS] | |||||||
| (e) | ||||||||
|---|---|---|---|---|---|---|---|---|
| LS | OS | Odds Ratio | Odds | Ratio | ||||
| Study or Subgroup | Events | Total | Events | Total | Weight | M-H, Fixed, 95% CI | M-H Fixed, | 95% CI |
| Donatini 2014 | 1 | 13 | 3 | 21 | 11.6% | 0.50 [0.05, 5.39] | ||
| Fossa 2013 | 8 | 17 | 11 | 15 | 33.8% | 0.32 [0.07, 1.43] | ||
| Lombardi 2012 | 1 | 30 | 7 | 126 | 14.2% | 0.59 [0.07, 4.95] | ||
| Mir 2013 | 18 | 18 | 26 | 26 | Not estimable | |||
| Wu 2018 | 1 | 21 | 2 | 23 | 9.9% | 0.53 [0.04, 6.25] | ||
| Zheng 2018 | 7 | 20 | 9 | 22 | 30.4% | 0.78 [0.22, 2.72] | ||
| Total (95% CI) | 119 | 233 | 100.0% | 0.54 [0.25, 1.17] | ||||
| Total events | 36 | 58 | ||||||
| Heterogeneity: Chi" = 0.79, df = | 4 (P = | 0.94); I | = 0% | 10 | ||||
| Test for overall effect: Z = 1.56 | (P = | 0.12) | 0.01 | 0.1 Favours [LS] | 1 Favours [OS] | |||
FIG. 5. Pooled estimate of secondary outcomes between laparoscopic surgery and open surgery: a operative time; b violation of capsules; c positive resection margin; d hospital stay; and e postoperative complications. LS laparoscopic surgery, OS open
This meta-analysis is the first to evaluate ACC survival rates based on the time interval. At 3 years, OS survival rates seem to be more favorable than LS, while the remainder of the time intervals demonstrated comparable survival rates between the two groups. Likewise, the HR of survival rates was similar between LS and OS, consistent with previous meta-analyses.14,37 Donatini et al.29 and Kastelan et al.25 both attributed the advancement in the technical expertise of laparoscopy and surgeon’s experi- ence as important factors for enhancing the role and status of LS in the management of ACC. It is widely acknowl- edged that given the rarity of ACC, the success of LS is already disadvantaged due to selection preference for OS in most institutions.25 Notably, many surgeons in the earlier phase of the learning curve avoid performing LS on larger tumors and patients with significantly elevated body mass index (BMI).38 Moreover, LS carries a higher risk of mortality and morbidity associated with the steep learning curve required for surgeons to determine the correct resection technique.25,39 Nonetheless, Porpiglia et al.32 postulated that LS is a reproducible surgical approach in ACC across various surgeons despite different levels of expertise within a single center. Furthermore, the LS approach seems to demonstrate comparable survival rates among the cohorts if surgical and operative parameters were kept consistent. Thus, evaluating the surgical
| (b) | ||||||||
|---|---|---|---|---|---|---|---|---|
| LS | OS | Weight | Odds Ratio | Odds | Ratio | |||
| Study or Subgroup | Events | Total | Events | Total | M-H, Fixed, 95% CI | M-H Fixed, | 95% CI | |
| Brix 2010 | 3 | 35 | 4 | 35 | 36.2% | 0.73 [0.15, 3.51] | ||
| Donatini 2014 | 0 | 13 | 1 | 21 | 11.1% | 0.51 [0.02, 13.36] | ||
| Fossa 2013 | 5 | 17 | 2 | 15 | 14.9% | 2.71 [0.44, 16.68] | ||
| Porpiglia 2010 | 0 | 18 | 0 | 25 | Not estimable | |||
| Wu 2018 | 7 | 21 | 6 | 23 | 37.8% | 1.42 [0.39, 5.20] | ||
| Total (95% CI) | 104 | 119 | 100.0% | 1.26 [0.55, 2.86] | ||||
| Total events | 15 | 13 | ||||||
| Heterogeneity: Chi = 1.48, df | = 3 (P = | 0.69); 12 | = 0% | 1 | ||||
| Test for overall effect: | Z = 0.55 | (P = | 0.59) | 0.01 | 0.1 | 10 | ||
| Favours [LS] | Favours [OS] | |||||||
100
| Mean | OS | Mean Difference IV, Fixed, 95% CI | |||||||
|---|---|---|---|---|---|---|---|---|---|
| (d) | LS | Weight | Mean Difference IV, Fixed, 95% CI | ||||||
| Study or Subgroup | Mean | SD | Total | SD | Total | ||||
| Donatini 2014 | 7 | 5 | 13 | 9 | 6 | 21 | 1.5% | -2.00 [-5.74, 1.74] | |
| Fossa 2013 | 6 | 6.5 | 17 | 13 | 5.25 | 15 | 1.2%-7.00 | [-11.08. - 2.92] | |
| Lombardi 2012 | 5.3 | 3.7 | 30 | 9.3 | 6.2 | 126 | 7.1% | -4.00 [-5.71 | |
| Mir 2013 | 4 1.2 | 18 | 6 | 1.625 | 26 | 29.6% | -2.00 -2.84, -1.16] | ||
| Wu 2018 | 6.3 | 2 | 21 | 8.6 | 4 | 23 | 6.1% | -2.30 [-4.15, -0.45] | |
| Zheng 2018 | 7 | 0.5 | 20 | 9.5 | 1.375 | 22 | 54.6% | -2.50 [-3.11, -1.89] | |
| Total (95% CI) | 119 | 233 | 100.0% | -2.49 [-2.95, -2.04] | |||||
| Heterogeneity: Chi | = 9.13, df = | 5 (P = 0.10); 12 = | 45% | 0 5 | |||||
| Test for overall effect: Z = 10.76 (P <0.00001) | -10 | Favours [LS] Favours [OS] | |||||||
10
approach, selection bias, and experience factors may be crucial when interpreting the survival rates between LS and OS.
Similar to overall survival rates, the recurrence-free rates showed comparable outcomes between the LS and OS cohorts, except for the 3-year interval, concurring with results reported in previous meta-analyses.37,40 While our findings showed no difference, the topic of recurrence remains an area of ongoing debate when considering the safety of LS in ACC. Historically, LS is associated with a higher risk of tumor capsule violation, which can elicit further complications such as peritoneal carcinomatosis. Furthermore, Leboulleux et al.12 theorized the type of surgical approach as the only risk factor during follow-up for the incidence of peritoneal carcinomatosis. Corre- spondingly, Cooper et al.28 reported that the LS group was more likely to develop recurrence of multifocal peritoneal carcinomatosis compared with the OS group. Conversely, Donatini et al.,29 Brix et al.,24 and Carnaille4 questioned the clarification of how the outcomes were reported with these earlier studies. Notably, selection bias and the sur- geon’s lack of experience were evident when performing LS as a surgical approach for ACC. As such, findings from the included studies24,25,32 in this meta-analysis, which showed similar recurrence-free rates between the two cohorts, allow the opportunity for further discussion points to be addressed on the safety of LS in managing ACC.
In addition to technical expertise, another major deter- minant for the selection of laparoscopic resection is based on tumor staging. Traditionally, LS is reserved for lesions smaller than 10 cm and without extra-adrenal involve- ment.24,29 Findings from our study revealed that the size of tumors resected in the LS group was relatively smaller than the OS group. As such, Kastelan et al.25 discussed that one of the inherent biases in their paper is the selection of LS for smaller tumors in favor of OS. However, Brix et al.24 matched patients based on tumor stage, and observed that OS and LS were comparable in recurrence-free survival rates for patients at AJCC stages I and II.41 Moreover, Tella et al.42 observed no difference in R0 margin status for tumor stages I-III, irrespective of whether LS or OS was performed in patients with ACC. While this is promising, these results warrant the attention of future studies to focus on individual tumor stages as an independent factor impacting the oncological outcomes. Comparatively, meta- analyses with subgroup analysis based on tumor staging have been achieved in other malignancies such as colon,45 gastric,44 gallbladder,45 and hepatocellular cancers.46 Therefore, extending these efforts into ACC can provide a more accurate representation of stage-dependent survival outcomes and recurrence rates between OS and LS.
In addition to tumor size, this meta-analysis demon- strated that survival rates with negative resection margin at 1, 2, 3, and 5 years were comparable between LS and OS. While the mechanisms related to the oncological feasibility of LS remain unanswered, Zheng et al.34 propose that the limited plane of view of surrounding tissues in the laparoscopic approach can lead to incomplete resection. As such, findings from their study were consistent with Cooper et al.,28 in showing the rate of positive resection margins being higher in LS compared with OS, whereas, from this meta-analysis, Brix et al.,24 Fossa et al.,30 and Mir et al.31 found no discrepancy in survival rates based on resection margins between the two cohorts. While these results are promising, only five studies reported these outcomes. Therefore, an increase in sample size with resection margin in future studies is required to help develop more accurate estimates.
The advantages of LS on perioperative outcomes in adrenal pathologies have been well-established in the lit- erature. Our study demonstrated that the length of hospital stay was shorter in the LS group, concordant with previous meta-analyses.37,40 Specifically, Donatini et al.29 men- tioned that when LS is correctly performed in ACC, patients have the potential to experience more favorable perioperative outcomes with faster recovery. Nonetheless, there was no difference in the operative time between cohorts, which is also consistent with previous findings in the literature.37 Contrary to earlier reports,12,47 the inci- dence of intraoperative capsular rupture was demonstrated
to be similar between the two groups in our meta-analysis. In addition to the capsular rupture, the rate of postoperative complications was also observed to be similar between the two groups, and this finding was in agreement with a previous literature review.37 Zheng et al.34 pointed out that the disadvantages of LS can be partly due to limitations of tactile sensation experienced with LS, which may increase the risk of tumor capsule rupture. Evidently, the perioper- ative benefits of LS in treating ACC from this meta- analysis largely support the growing trends in the use of laparoscopic methods. Furthermore, in attempts to demonstrate the feasibility of LS, more studies that report perioperative outcomes may help efforts to consider LS as an alternative treatment therapy in managing ACC.
One of the critical limitations of this meta-analysis is in relation to the design of the studies, which mainly con- sisted of retrospective cohort studies. Additionally, the majority of studies were conducted in a single-center set- ting, thus analyzing with additional multicenter studies could improve the sample size and generalizability of this meta-analysis. Since many different types of pharmaco- logical interventions are administered to patients with ACC, it was difficult to homogenize the adjuvant therapies given. Therefore, the difference in these interventions could have ultimately contributed to heterogeneity in our outcomes. Similarly, it is challenging to account for the differences between centers in how they regulate protocols, choice of equipment, surgical technique, and surgeons’ experience, therefore also influencing heterogeneity. Lastly, there was considerable heterogeneity in the out- comes, such as recurrence-free rates at 5 and 10 years, 1-year survival rate with a negative resection margin, incidentaloma, tumor size, Weiss score, and operative time. Despite these limitations, this meta-analysis demonstrated OS should still be considered as the standard operative approach; however, LS could potentially be an alternative option in managing ACC in selected cases with appropriate laparoscopic expertise. As such, it would be beneficial for future studies to address these limitations in continuing their efforts in evaluating the safety and efficacy of LS and OS in the management of ACC.
CONCLUSIONS
In comparison with LS, OS should be considered the standard operative approach in managing patients with ACC for showing more favorable trends in overall survival and recurrence-free rates. These trends are likely due to a lower rate of positive resection margin compared with LS. However, LS showed the potential in managing selected ACC cases with appropriate laparoscopic expertise. Despite the promising results in this meta-analysis, further
studies based on tumor stage- and resection margin-de- pendent survival outcomes are required to investigate the differences between LS and OS. Moreover, long-term randomized controlled trials are necessary to help further strengthen the role of LS in the management of ACC.
Supplementary Information The online version contains supplementary material available at https://doi.org/10.1245/s10434- 022-12711-w.
ACKNOWLEDGMENT The authors would like to thank M.L.S. Larry Prokop for the literature search.
AUTHOR CONTRIBUTIONS HN, SM, and HC conceived and designed the study; reviewed the literature; collected, analyzed, and interpreted the data; and drafted and critically revised the manuscript. HN, SM, RW, JEH, NEG, FAA, RHM, CAT, BEJ, and HC reviewed the literature; collected, analyzed, and interpreted the data; and drafted the manuscript. All authors read and approved the final manuscript.
FUNDING This research did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.
DATA AVAILABILITY The data set used for this meta-analysis will be shared upon request to the study authors.
DISCLOSURE Benjamin Johnson has a financial relationship with Intuitive Surgical for proctoring and educational events, however these were not related to this current paper. Hayato Nakanishi, Shahid Miangul, Rongzhi Wang, Joe El Haddad, Nour El Ghaza, Fatma Ali Makhzoum Abdulsalam, Reem H. Matar, Christian A. Than, and Herbert Chen declare they have no competing interests.
ETHICAL APPROVAL This systematic review and meta-analysis does not require ethical approval.
REFERENCES
1. Sharma E, Dahal S, Sharma P, et al. The characteristics and trends in adrenocortical carcinoma: a United States population based study. J Clin Med Res. 2018;10(8):636-40.
2. Ayala-Ramirez M, Jasim S, Feng L, et al. Adrenocortical carci- noma: clinical outcomes and prognosis of 330 patients at a tertiary care center. Eur J Endocrinol. 2013;169(6):891-9.
3. Marincola Smith P, Kiernan CM, Tran TB, et al. Role of addi- tional organ resection in adrenocortical carcinoma: analysis of 167 patients from the U.S. adrenocortical carcinoma database. Ann Surg Oncol. 2018;25(8):2308-15.
4. Carnaille B. Adrenocortical carcinoma: Which surgical approach? Langenbecks Arch Surg. 2012;397(2):195-9.
5. Smith CD, Weber CJ, Amerson JR. Laparoscopic adrenalectomy: new gold standard. World J Surg. 1999;23(4):389-96.
6. Wang HS, Li CC, Chou YH, Wang CJ, Wu WJ, Huang CH. Comparison of laparoscopic adrenalectomy with open surgery for adrenal tumors. Kaohsiung J Med Sci. 2009;25(8):438-44.
7. Yan Y, Cheng J, Chen K, Liu TF, Ning G. Better clinical benefits and potential cost saving of an enhanced recovery pathways for laparoscopic adrenalectomy. Gland Surg. 2022;11(1):23-34.
8. Hazzan D, Shiloni E, Golijanin D, Jurim O, Gross D, Reissman P. Laparoscopic vs open adrenalectomy for benign adrenal neo- plasm. Surg Endosc. 2001;15(11):1356-8.
9. Pigg RA, Fazendin JM, Porterfield JR, Chen H, Lindeman B. Patient satisfaction is equivalent for inpatient and outpatient minimally-invasive adrenalectomy. J Surg Res. 2022;269:207-11.
10. Miller BS, Gauger PG, Hammer GD, Doherty GM. Resection of adrenocortical carcinoma is less complete and local recurrence occurs sooner and more often after laparoscopic adrenalectomy than after open adrenalectomy. Surgery. 2012;152(6):1150-7.
11. Kebebew E, Siperstein AE, Clark OH, Duh QY. Results of laparoscopic adrenalectomy for suspected and unsuspected malignant adrenal neoplasms. Arch Surg. 2002;137(8):948-51 (discussion 952-943).
12. Leboulleux S, Deandreis D, Al Ghuzlan A, et al. Adrenocortical carcinoma: Is the surgical approach a risk factor of peritoneal carcinomatosis? Eur J Endocrinol. 2010;162(6):1147-53.
13. Lombardi CP, Raffaelli M, De Crea C, Bellantone R. Role of laparoscopy in the management of adrenal malignancies. J Surg Oncol. 2006;94(2):128-31.
14. Langenhuijsen J, Birtle A, Klatte T, Porpiglia F, Timsit MO. Surgical management of adrenocortical carcinoma: impact of laparoscopic approach, lymphadenectomy, and surgical volume on outcomes-a systematic review and meta-analysis of the current literature. Eur Urol Focus. 2016;1(3):241-50.
15. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097.
16. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta- analyses. Eur J Epidemiol. 2010;25(9):603-5.
17. Fassnacht M, Johanssen S, Quinkler M, et al. Limited prognostic value of the 2004 International Union Against Cancer staging classification for adrenocortical carcinoma: proposal for a Revised TNM Classification. Cancer. 2009;115(2):243-50.
18. Lughezzani G, Sun M, Perrotte P, et al. The European Network for the Study of Adrenal Tumors staging system is prognostically superior to the international union against cancer-staging system: a North American validation. Eur J Cancer. 2010;46(4):713-9.
19. Hermanek P, Wittekind C. The pathologist and the residual tumor (R) classification. Pathol Res Pract. 1994;190(2):115-23.
20. Guyot P, Ades AE, Ouwens MJ, Welton NJ. Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol. 2012;12:9.
21. Higgins JPT, Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions. 2nd edn. Hoboken: Wiley; 2020.
22. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557-60.
23. Sterne JA, Egger M. Funnel plots for detecting bias in meta- analysis: guidelines on choice of axis. J Clin Epidemiol. 2001;54(10):1046-55.
24. Brix D, Allolio B, Fenske W, et al. Laparoscopic versus open adrenalectomy for adrenocortical carcinoma: surgical and onco- logic outcome in 152 patients. Eur Urol. 2010;58(4):609-15.
25. Kastelan D, Knezevic N, Zibar Tomsic K, et al. Open vs laparoscopic adrenalectomy for localized adrenocortical carci- noma. Clin Endocrinol (Oxf). 2020;93(4):404-8.
26. Tseng J, Diperi T, Gonsalves N, et al. Operative approach and case volume are associated with negative resection margins for adrenocortical carcinoma. Surg Endosc. 2022. https://doi.org/10. 1007/s00464-022-09167-0.
27. Lombardi CP, Raffaelli M, De Crea C, et al. Open versus endoscopic adrenalectomy in the treatment of localized (stage
I/II) adrenocortical carcinoma: results of a multiinstitutional Italian survey. Surgery. 2012;152(6):1158-64.
28. Cooper AB, Habra MA, Grubbs EG, et al. Does laparoscopic adrenalectomy jeopardize oncologic outcomes for patients with adrenocortical carcinoma? Surg Endosc. 2013;27(11):4026-32.
29. Donatini G, Caiazzo R, Do Cao C, et al. Long-term survival after adrenalectomy for stage I/II adrenocortical carcinoma (ACC): a retrospective comparative cohort study of laparoscopic versus open approach. Ann Surg Oncol. 2014;21(1):284-91.
30. Fossa A, Rosok BI, Kazaryan AM, et al. Laparoscopic versus open surgery in stage I-III adrenocortical carcinoma-a retro- spective comparison of 32 patients. Acta Oncol. 2013;52(8):1771-7.
31. Mir MC, Klink JC, Guillotreau J, et al. Comparative outcomes of laparoscopic and open adrenalectomy for adrenocortical carci- noma: single, high-volume center experience. Ann Surg Oncol. 2013;20(5):1456-61.
32. Porpiglia F, Fiori C, Daffara F, et al. Retrospective evaluation of the outcome of open versus laparoscopic adrenalectomy for stage I and II adrenocortical cancer. Eur Urol. 2010;57(5):873-8.
33. Wu K, Liu Z, Liang J, et al. Laparoscopic versus open adrenalectomy for localized (stage 1/2) adrenocortical carcinoma: experience at a single, high-volume center. Surgery. 2018;164(6):1325-9.
34. Zheng GY, Li HZ, Deng JH, Zhang XB, Wu XC. Open adrenalectomy versus laparoscopic adrenalectomy for adreno- cortical carcinoma: a retrospective comparative study on short- term oncologic prognosis. Onco Targets Ther. 2018;11:1625-32.
35. Margonis GA, Kim Y, Prescott JD, et al. Adrenocortical carci- noma: impact of surgical margin status on long-term outcomes. Ann Surg Oncol. 2016;23(1):134-41.
36. Grubbs EG, Callender GG, Xing Y, et al. Recurrence of adrenal cortical carcinoma following resection: surgery alone can achieve results equal to surgery plus mitotane. Ann Surg Oncol. 2010;17(1):263-70.
37. Hu X, Yang WX, Shao YX, Dou WC, Xiong SC, Li X. Mini- mally invasive versus open adrenalectomy in patients with adrenocortical carcinoma: a meta-analysis. Ann Surg Oncol. 2020;27(10):3858-69.
38. Yeo CT, Pasieka JL, Assadzadeh GE, Harvey A. The adoption of the endoscopic retroperitoneal approach to the adrenal gland- conversion factors and learning points. Am J Surg. 2022;224(2):703-9.
39. Fazendin J, Chen H, Lindeman B. Influence of fellowship edu- cational experience on practice patterns for adrenalectomy: a
survey of recent AAES fellowship graduates. Am J Surg. 2021;221(3):626-30.
40. Autorino R, Bove P, De Sio M, et al. Open versus laparoscopic adrenalectomy for adrenocortical carcinoma: a meta-analysis of surgical and oncological outcomes. Ann Surg Oncol. 2016;23(4):1195-202.
41. Amin MB, Greene FL, Edge SB, et al. The eighth edition AJCC cancer staging manual: continuing to build a bridge from a population-based to a more “personalized” approach to cancer staging. CA Cancer J Clin. 2017;67(2):93-9.
42. Tella SH, Kommalapati A, Yaturu S, Kebebew E. Predictors of survival in adrenocortical carcinoma: an analysis from the national cancer database. J Clin Endocrinol Metab. 2018;103(9):3566-73.
43. Theophilus M, Platell C, Spilsbury K. Long-term survival fol- lowing laparoscopic and open colectomy for colon cancer: a meta-analysis of randomized controlled trials. Colorectal Dis. 2014;16(3):O75-81.
44. Vinuela EF, Gonen M, Brennan MF, Coit DG, Strong VE. Laparoscopic versus open distal gastrectomy for gastric cancer: a meta-analysis of randomized controlled trials and high-quality nonrandomized studies. Ann Surg. 2012;255(3):446-56.
45. Nakanishi H, Miangul S, Oluwaremi TT, Sim BL, Hong SS, Than CA. Open versus laparoscopic surgery in the management of patients with gallbladder cancer: a systematic review and meta- analysis. Am J Surg. 2022;224(1 Pt B):348-57.
46. Tsai KY, Chen HA, Wang WY, Huang MT. Long-term and short- term surgical outcomes of laparoscopic versus open liver resec- tion for hepatocellular carcinoma: Might laparoscopic approach be better in early HCC? Surg Endosc. 2019;33(4):1131-9.
47. Miller BS, Ammori JB, Gauger PG, Broome JT, Hammer GD, Doherty GM. Laparoscopic resection is inappropriate in patients with known or suspected adrenocortical carcinoma. World J Surg. 2010;34(6):1380-5.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.