Laparoendoscopic single-site nephrectomy versus conventional laparoendoscopic nephrectomy for kidney tumor: a systematic review and meta-analysis

With the increasing application of laparoendoscopic single-site nephrectomy (LESS-N) in kidney tumor, accumulating studies compared it with conventional laparoendoscopic nephrectomy (CL-N). However, controversial outcomes were reported. Hence, this meta-analysis was carried out to clarify these issues. Online databases PubMed, EMBASE and the Cochrane Library were searched comprehensively for eligible studies published before 24 July 2018. Odds ratios (ORs) or standardized mean differences (SMDs) with corresponding 95% confidence intervals (CIs) were collected for evaluating the pooled results of relevant outcomes. Ultimately, 13 eligible articles were enrolled. Meanwhile, compared with CL-N, LESS-N was related to a longer operation time (SMD: 0.40; 95% CI, 0.23–0.58; P=0.000), a shorter length of hospital stay (LOS) (SMD: −0.32; 95% CI, −0.62 to −0.02; P=0.034), a lower visual analog scale (VAS) score (SMD: −0.89; 95% CI, −1.22 to −0.56; P=0.000) and a lower analgesic requirement (SMD: −0.55; 95% CI, −0.87 to −0.23; P=0.001). There was no statistical difference in the postoperative day of oral intake, estimated blood loss (EBL), conversion rate, perioperative complications, intraoperative complications, postoperative complications, minor complications and major complications between LESS-N and CL-N. Patients with LESS-N for kidney tumor could have a longer operation time and shorter LOS, and meanwhile could need less analgesics and suffer less pain after LESS-N.


Introduction
Over the last decades, open nephrectomy for kidney tumor has been gradually replaced by laparoendoscopic nephrectomy first reported by Clayman et al. in 1991 [1] and it has been demonstrated to gain great advantages in cosmetic appearance, postoperative pain, hospital stay and perioperative complications [2,3]. Recently, new less invasive technologies have been applied in treatment including Natural orifice transluminal endoscopic surgery (NOTES) [4] and laparoendoscopic single-site (LESS) surgery [5]. However, NOTES needs special equipment and more technical skills, which limits its application [4]. As a result, LESS has gained a lot of attention since it was first reported in Urology in 2007 [5].
As for LESS nephrectomy (LESS-N), it is a kind of minimally invasive surgery based on LESS, which is performed through a single incision, usually around the umbilical or transperitoneal region. Sometimes, an additional 3-mm trocar used in right kidney nephrectomies for liver retraction is also considered as LESS-N [5,6]. In terms of conventional laparoendoscopic nephrectomy (CL-N), it is also a minimally invasive surgery accomplished through usually three keyholes. Nowadays, more and more researches focus on the comparison of LESS-N and CL-N, including operation time, estimated blood loss (EBL), postoperative day of oral intake, length of hospital stay (LOS), visual analog scale score (VAS), analgesic requirement, conversion rates, perioperative complications, intraoperative complications, postoperative complications, minor complication and major complications. However, most of them are small series, even with conflicting results.
As a powerful tool, meta-analysis could provide more reliable results than a single study by combining all eligible studies, especially in explaining controversial conclusions. Therefore, we systematically and comprehensively searched eligible articles and evaluated their potential efficiency, safety and advantages of LESS-N in comparison of CL-N.

Literature search
PubMed, EMBASE and the Cochrane Library were searched comprehensively for eligible studies published before 24 July 2018. The search strategy consisted of the following keywords in combination with Medical Subject Headings (MeSH) terms and text words: ('single site laparoscopy/laparoendoscopy' or 'single port laparoscopy/laparoendoscopy') and ('conventional laparoscopy/laparoendoscopy' or 'traditional laparoscopy/laparoendoscopy') and 'nephrectomy' . Meanwhile, additional articles were searched in the database manually, when we searched relevant reviews and the reference list of original articles.

Inclusion and exclusion criteria
The eligible studies needed to meet the following inclusion criteria: (1) studies comparing LESS-N with CL-N for kidney tumor; (2) reporting at least one of the following perioperative outcomes: operation time, EBL, postoperative day of oral intake, LOS, VAS, analgesic requirement, conversion rate, perioperative complications, intraoperative complications, postoperative complications, minor complication or major complications; (3) sufficient data could be extracted from the enrolled studies. Additionally, the exclusion criteria included the following points: (1) not meet the inclusion criteria; (2) reviews, conference meeting abstracts, case reports or comments; (3) data could not be extracted; (4) overlapping data from the same institution.

Data extraction and study quality
The preferred reporting items for systematic review and meta-analysis (PRISMA) was utilized for reporting this article [7]. Two independent authors were responsible for extracting data from the included studies. Demographic characteristics were compared including age, gender ratio, body mass index (BMI), side of procedure, tumor size, American Society of Anesthesiologists score (ASA), history of prior abdominal surgery. The perioperative outcomes were collected including operation time, EBL, postoperative day of oral intake, LOS, VAS, analgesic requirement, conversion rate, perioperative complications, intraoperative complications, postoperative complications, minor complication and major complications. Postoperative complications were graded according to the Clavien-Dindo system [8]. The evidence level of included studies was rated by one author according to the criteria provided by the Oxford Centre for Evidence-Based Medicine. The methodological quality of randomized controlled trial (RCT) was evaluated by the Jadad scale [9]. Meanwhile, the methodological quality of the retrospective studies was assessed by the modified Newcastle-Ottawa scale [10]. Stata 14.0 was utilized to perform this meta-analysis. All statistical methods met the principles mentioned in PRISMA. The standardized mean differences (SMDs) or odds ratios (ORs) were used for continuous and dichotomous variables, respectively; 95% confidence intervals (CIs) were reported in all results. Studies presenting continuous data as means and range values, the standard deviations were calculated using the methods described by Hozo et al. [11]. Statistical heterogeneity was evaluated by chi-square test and inconsistency (I 2 ). If high heterogeneity was tested, a random-effect model was used. Otherwise, a fixed-effect model would be applied. The pooled effects were tested by the Z-test and the results with P<0.05 were considered to be statistically significant. Contour-enhanced funnel plots and L'abbe graph were used to evaluate publication bias for dichotomous variables and continuous variables, respectively. Abbreviations: R, retrospective; RP, retrospective design, prospective data collection collection. Matching: 1 = age, 2 = gender ratio, 3 = BMI, 4 = side of procedure, 5 = ASA, 6 = history of prior abdominal surgery, 7 = tumor size, 8 = same surgeon. 1 No single-port access device was used; the adjacent trocars were inserted through a single incision. 2 The score of the Jadad scale for the methodological quality of the RCTs.

Publication bias
In this meta-analysis, contour-enhanced funnel plots were used for dichotomous variables (Figures 6 and 8). L' Abbe graphs were used for continuous variables (Figures 7 and 9). No publication bias was found in our study.

Discussion
This meta-analysis compared not only LESS-N with CL-N, but also LESS-RN with CL-RN for kidney tumor and the pooled results analyzed in this article, showed that LESS-N was related to a longer operation time, a shorter LOS, a lower VAS and a lower analgesic requirement. Moreover, LESS-N had similar perioperative outcomes with CL-N including postoperative day of oral intake, EBL, conversion rate, perioperative complications, intraoperative complications, postoperative complications, minor complications and major complications. As for the analysis of LESS-RN and CL-RN, there were no significant differences compared with the analysis of LESS-N and CL-N.
In this article, we found operation time of LESS-N was longer than CL-N. LESS operation is difficult and requires technical proficiency, laparoendoscopic operation skills and high capability of handling emergency. Surgeons contacting LESS-N in the early stage are hard to master this technology quickly [15]. The learning curve of LESS-N is quite long, which could be the main reason for longer operation time for LESS-N. Park et al. [22] reported that operation time decreased with the increasing number of patients. So extensive training in LESS-N will reduce operation time. Meanwhile, proper devices for LESS-N could make the operation easier. Some institutions utilize homemade devices to obtain satisfactory outcomes [13,15,[17][18][19]22,25]. Some others use special devices like OCTO Port, SILS-Port and GelPort etc [12][13][14]16,[22][23].
The safety and feasibility of LESS-N is of great concern. Special devices and related training applied made sure surgeons could perform LESS-N as well as CL-N. Conversion rate was a good indicator for evaluating whether the surgery is safe and feasible. Conversion into open surgery or conventional laparoscopy surgery was inevitable for some situations such as bleeding etc. The pooled data showed conversion rate of LESS-N and CL-N was 2.6 and 3.0%, respectively. Meanwhile no difference in conversion rate was found between two groups. The incidence of complications is widely recognized as an important indicator of surgical safety. The Clavien-Dindo system classified perioperative complications into five grades, in which grades 1 and 2 are considered as minor complications, meanwhile grades 3, 4 and 5 are thought as major complications [8]. Complications after nephrectomy were dominantly minor complications, which included fever, diarrhea, ileus, bleeding etc. Major complications such as hydronephrosis, severe pneumonia, intensive care after surgery rarely happened [19][20][21][22]. In this meta-analysis, there is also no significant difference in perioperative complications, intraoperative complications, postoperative complications, major complications and minor complications between LESS-N and CL-N.
Results from this meta-analysis demonstrated lower VAS and analgesic requirement in LESS-N group, which was similar to a previously published meta-analysis [26] showing that patients with LESS-N suffer less postoperative pain than patients with CL-N. LESS-N and CL-N are considered as minimally invasive surgeries, and the main difference between the two surgical procedures is the number of ports applied. Single one incision with less muscle-splitting in LESS-N is considered a main reason [16,21]. In CL-N, removing specimen needs a corresponding incision. However, surgical procedure and removal of specimen are operated through one incision in LESS-N, which also reduces surgical scar and leads to a better cosmetic. Several studies [27][28][29] reported morcellating specimen may also decrease postoperative discomfort. However, morcellation might induce tumor seeding, which limits its use in surgery [16]. Some researchers studied warm ischemic time, which affects renal function after nephrectomy by influencing blood supply of renal parenchyma. Shin et al. [19] reported LESS-PN had a longer mean warm ischemic time than CL-PN (19.8 + − 13.1 vs. 26.5 + − 10.5 min, P=0.001). However, postoperative renal function, evaluated by preoperative to postoperative eGFR change, was only found significantly higher in CL-PN than LESS-PN on postoperative day 1 but there was no difference between CL-PN and LESS-PN on postoperative days 0, 1, 7, and at 3 and 6 months, demonstrating warm ischemic time primarily affected renal function within a short period after nephrectomy. Meanwhile, Springer et al. [30] reported percent of eGFR preserved affected postoperative renal function primarily other than warm ischemic time. So size and complexity of tumor determining percent of parenchyma preserved dominantly indicates renal function postoperatively.
The meta-analysis has some limitations that should not be ignored. The main limitation is that nearly all the studies included were retrospective, except one prospective non-randomized study and one RCT. Besides, most studies are small sample studies. Second, surgeons with different surgical experience used diverse devices. The background of each study was different. But the results of the LESS-RN and CL-RN are the same with the results of the LESS-N and CL-N. In the end, all studies lack long period of follow-up, which means that recurrence, metastasis and mortality could not be evaluated. So long-term follow-up between LESS-N and CL-N is really needed. Nevertheless, this meta-analysis provides latest information on comparing LESS-N and CL-N for kidney tumor. The conclusion is quite convincing because of a large number of patients. We not only analyzed the perioperative outcomes of LESS-N and CL-N, but also the perioperative outcomes of LESS-RN and CL-RN. Different ways to evaluate methodological quality of included studies were applied. And contour-enhanced funnel plots and L' Abbe graphs were applied in this meta-analysis showing no publication bias.

Conclusions
Taken together, our results shed light on that patients with LESS-N for kidney tumor could have a longer operation time and shorter LOS, and meanwhile could need less analgesics and suffer less pain after LESS-N. Due to the aforementioned limitations, larger samples of more strictly designed RCTs are required to verify our findings.