Methods
Search Strategy and Study Selection
This study followed the PRISMA guidelines (10). Published studies which investigated LPD versus OPD for PDCA were systematically searched in PubMed, Web of Science, EMBASE, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov databases before March 22, 2021, by two independent researchers (QF, WL). The combinations of the following key terms and their combinations were used: laparoscopic, open, conventional, whipple, and pancreaticoduodenectomy. In order to gain additional studies, the references of eligible studies were manually searched.
Inclusion and Exclusion Criteria
Two investigators (QF, WL) reviewed currently available literature and screened all titles and abstracts independently and identified eligible studies according to the following criteria.
The inclusion criteria were as follows: 1) participants: patients with PDAC; 2) types of interventions: LPD and OPD; 3) types of studies: randomized controlled trials (RCTs), retrospective studies, cohort studies, and case–control studies; and 4) primary articles published in English.
The exclusion criteria were as follows: 1) non-English or experimental studies; 2) studies without available data; and 3) the publication type was editorial, abstract, letter, case report, and expert opinion.
Data Extraction and Quality Assessment
The original data from all candidate articles were independently assessed and extracted by two reviewers (QF, WL) by using a unified datasheet, and any ambiguity was resolved by a third researcher (ZX) who was consulted to review the study to reach a consensus. Data extracted included the following items: study and patient characteristics and operative and postoperative outcomes. Study and patient characteristics include first author, country, publication year, research design, sample size, and mean age; the latter included operative time, blood loss, blood transfusion, overall morbidity and 30-day mortality, length of hospital stay (LOS), R0 resections, number of harvested lymph nodes, overall survival (OS), postoperative long-term survival, and the time to starting postsurgical adjuvant chemotherapy. The quality of the included studies was assessed by the Newcastle-Ottawa Scale (NOS) (11). Every included study was independently evaluated by two authors (QF, WL), and NOS score ≥6 is considered as being of high quality.
Statistical Analysis
Review Manager 5.3 software was used to analyze the data. The weighted mean difference (WMD) with the 95% confidence interval (CI) and odds ratio (OR) were used to compare continuous variables and dichotomous, respectively. For overall survival data, the method of Tierney et al. was adopted to estimate natural logarithm hazard ratios (HRs), standard errors (SE), and 95% CIs from survival curves (12). We adopted the method described by Hozo et al. to convert medians with ranges into means with standard deviations (13). Begg’s funnel plot and Egger’s test were used to assess potential publication bias. Statistical heterogeneity was quantified using Higgin’s I2 index. A fixed-effects model (FEM) was adopted when heterogeneity is low or moderate (I2 < 50%), while when heterogeneity is high (I2 ≥ 50%), a random-effects model (REM) will be used.
Discussion
Since 1994, the first case of LPD was performed by Gagner et al., and the magnifying effect and visual field advantage of laparoscopic technique made it popular. Almost a decade later, Giulianotti et al. performed the first robotic pancreaticoduodenectomy (RPD) in Italy in 2003. RPD and LPD are both minimally invasive treatments for pancreatic and periampullary malignancies and some benign diseases. At present, with the development of laparoscopic surgical instruments and the accumulation of surgical experience, LPD has been proven to be safe and feasible and has many advantages over OPD, but the operation is more difficult and time-consuming (24, 25). Owing to the complicated operation process and high requirements for surgeons, LPD is considered as the “Everest” of abdominal minimally invasive surgery. Many institutions choose to include patients with benign diseases when they carry out LPD in the initial stage. In order to compare the real difference between LPD and OPD in the treatment of PDAC, we analyzed the data from the literature that cases were pathologically diagnosed as PDAC. Finally, the present meta-analysis included the latest studies from 2014 to 2021 to compare the safety and efficacy of LPD and OPD for PDAC. Although none of the studies was RCTs, most of included studies were relatively high quality according to NOS. Finally, our study included 10 publications with 11,535 (LPD 1,514 vs. OPD 10,021) patients and reflects the newest surgical results for the treatment of PDAC.
Three previous meta-analyses comparing perioperative and oncologic outcomes of LPD to OPD were published in the last 3 years. However, two of them (7, 9) included an article about RPD in the treatment of PDAC. The study of Chen et al. (7) is the only meta-analysis that focuses on LPD and OPD. Six articles were included in the study of Chen et al. and focused on oncologic outcomes and long-term survival. They found that LPD was associated with longer 3-year survival, 4-year survival, and 5-year survival time, and there was no difference in 1-year survival, 2-year survival time, lymph nodes harvested, the number of positive lymph nodes, and R0 rate. The meta-analysis of Yin et al. (9) covered 9,144 PDAC participants from six retrospective studies, and the meta-analysis of Sun et al. (8) covered 1,377 minimally invasive pancreaticoduodenectomy (LPD and RPD) and 9,865 OPD from nine retrospective studies. They advised that LPD was associated with higher R0 rate, longer operative time, fewer blood loss, lower transfusion rate, and shorter length of hospital stay. There were no significant differences in morbidity, POPF, DGE, the number of harvested lymph nodes, and mortality.
Compared with the results of other studies, our study included some recent studies and excluded the study comparing RPD and OPD. The results of our meta-analysis showed LPD has a shorter LOS and less blood loss but no difference in blood transfusion. Why do we have different results for blood loss and the same blood transfusion rate in the two groups? It is because five studies with a total of 1,195 patients (14, 15, 18, 22, 23) had reported blood loss. Another five studies (14, 18, 20, 22, 23) with 1,260 patients reported blood transfusion, and heterogeneity was very high (I2 = 81%), indicating that there is obvious publication bias. The results of our meta-analysis showed LPD has a longer operative time than OPD, which was similar with the study of Yin et al. The main factors that lead to the longer operation time of LPD are longer pancreatectomy and digestive tract reconstruction under a laparoscope.
The current meta-analysis shows non-significant difference in the 30-day mortality, overall complication rates, POPF, and the incidence of severe complications (Clavien–Dindo 3/4 grade complications) between the two groups, indicating that the safety of the two groups was similar. PPH is one of the most severe complications after pancreatic surgery. The meta-analysis of Floortje van Oosten et al. (26) showed that the incidence of PPH was about 5% after pancreatectomy, and the overall mortality caused by PPH accounted for 21%. Our meta-analysis shows that the rate of PPH was 6% (60/952 patients; range 4.1%–10.7%), and pooling the data of five studies (14, 16, 18, 20, 23) with 952 patients revealed no difference in PPH between the LPD and OPD groups.
Negative margin and the number of lymph node dissection are two important malignancy prognosis factors in PD. Pooled data from this meta-analysis revealed that LPD has a higher rate of R0 resection than OPD. We think that this may be explained by the fact that patients with PDAC in early stage were selected to perform LPD. From the perspective of tumor radical effect, the results of this study show that the two surgical methods have the same effect in the number of lymph node dissection, suggesting that LPD and OPD have the same tumor radical effect, which is basically consistent with the results of most existing clinical studies.
When it comes to long-term survival, according to our search, there are still no RCTs comparing the long-term survival between LPD to OPD in patients with PDAC. At present, the largest overall survival outcomes data of LPD in the treatment of PDAC come from the National Cancer Database (NCDB). Kantor et al. reported that 8,213 patients with PDAC underwent PD (828 underwent LPD and 7,385 underwent OPD) and revealed a non-significant difference in survival time in the two groups (20.7 vs. 20.9 months) (19). However, Chapman et al. compared the survival data of 1,768 patients with PDAC (248 underwent LPD and 1,520 underwent OPD) also from NCDB and suggested that LPD and OPD can achieve a median overall survival of 19.8 and 15.6 months, respectively (p = 0.022) (21). Although our meta-analysis revealed that the LPD group has an earlier time to get starting adjuvant chemotherapy after surgery, there is no significant difference in OS (HRs: 1.30; 95% CI 0.80 to 2.13; p = 0.41). In some ways, the pooled data demonstrated that LPD is not oncologically inferior to OPD and even can achieve superior oncologic outcome compared with OPD.
To evaluate the safety and efficiency of LPD for PDAC, this meta-analysis included 10 studies and revealed that LPD was comparable to OPD. But some limitations in this study should be considered. First, no RCTs were included which may contribute selection bias. Furthermore, of the 10 included studies, TNM stage, tumor size, and differentiation degree of patients with PDAC have not been reported in some studies. What is more, only a few studies reported long-term survival outcomes such as overall survival and 5-year survival time of LPD. Therefore, further large-scale prospective comparative studies and RCTs are expected to provide more convincing results to further assess the effectiveness and safety of LPD for patients with PDAC.