Introduction

Ureteropelvic junction obstruction (UPJO) is the most common congenital malformation in children with urinary tract obstruction and is one of the common causes of hydronephrosis in children [1]. With the application and popularization of prenatal ultrasonography, more and more asymptomatic hydronephrosis has been detected. Some children with hydronephrosis gradually aggravated, seriously affecting renal function. For UPJO patients who meet the surgical indications, surgical treatment is recommended [2]. The ureteric double-J stent is a minimally invasive procedure that provides an alternative to open surgery in infants with primary hydronephrosis [3]. The purpose is to relieve the obstruction, reconstruct the dilated renal pelvis, and protect renal function from deterioration.

Prenatal fetal hydronephrosis is discovered by ultrasound at a rate of 1% to 2%, which has led to the timely discovery of congenital hydronephrosis in an increasing percentage of children [4], with physiological pelvic dilation occurring after birth With the growth and development of newborns, after follow-up and observation, With the growth and development of babies, after follow-up and observation, it can relieve itself and be managed conservatively after a period of time, However, 25 to 30% of prenatal hydronephrosis has obstruction at the ureteropelvic junction, hydronephrosis is unable to heal itself or disappear over time, urine outflow is obstructed, accumulation in the renal pelvis and calyces, leading to increased pressure in the collecting system, and affects renal blood flow, leading to impaired renal function [5].

When compared to traditional open pyeloplasty (OP), laparoscopic pyeloplasty (LP) offers better recovery and less issues [6]. Laparoscopic pyeloplasty has gradually become another option for the treatment of UPJO. However, for children with small abdominal space, endoscopic suture and knot tying is more difficult for surgeons, requiring robotic surgery. Since the emergence of the robotic platform, robot-assisted laparoscopic pyeloplasty (RP) has been rapidly developed in the field of pediatric urology surgery with its improved ergonomics, wristed instruments, and 3-D magnified vision, and it has been rapidly accepted by pediatric urologist’s surgeons [7].

Robot-assisted laparoscopic pyeloplasty is a relatively new technique that may have advantages or disadvantages over previous classical surgical methods. As with any new healthcare technology, RP requires rigorous evaluation. Individual studies and reviews to date have provided scant evidence on the clinical efficacy of RP compared with LP. A few early studies reported that the average operation time was shorter in RP than in LP [8]. In recent years, researchers have shown that RP has a longer operation time than LP [9]. One issue when evaluating the complication rate of pyeloplasty is the existing variability among the different hospitals, which can have a negative impact on comparisons between studies [10]. In terms of the treatment effect of RP and LP in children with hydronephrosis, both can achieve satisfactory results, and the cure rate can reach more than 90%, but the cost of RP is significantly higher than that of LP [11]. However, some studies have shown that the length of hospital stay in the RP group is shorter than that in LP group, and RP has a greater advantage over than LP group [12]. Whether RP has a greater advantage in the treatment of UPJO in children needs to be confirmed by studies with a larger sample size and longer follow-ups. We performed meta-analysis to comprehensively evaluate the effects of RP and LP in the treatment of UPJO, to help patients and clinicians make decisions.

Methods

Search methods for identification of studies

English databases (PubMed, Cochrane Database, Web of Science, Ovid, and Embase) and Chinese databases (CNKI, VIP, CBM, Wanfang) were searched independently by two searchers (CZM and WCH). We selected “Laparoscopic pyeloplasty”, “Ureteropelvic junction obstruction”, and “Robotic pyeloplasty” as the keywords in this literature search, and relevant free words were searched in the database, using the form of joint search of subject terms and free words in each database. We performed the original searches from inception to 30 June 2022. The search strategy is listed in Tables 1 and 2.

Table 1 Search strategy
Full size table
Table 2 The main characteristics of the included studies
Full size table

Criteria for considering studies for this review

Two researchers (HSJ and KZ) independently reviewed all records to assess literature eligibility, and inconsistencies were resolved through open discussion.

Inclusion criteria were: (1) studies focused on children (< 18 years old) with a diagnosis of ureteropelvic junction obstruction; (2) undergoing robotic pyeloplasty; (3) in which laparoscopic pyeloplasty was performed as a comparator; (4) evaluating one or more of the following outcomes: primary outcomes (success rate, complication rate), secondary outcomes (rate of conversion to open surgery, duration of hospital stay, operative time); (5) We included randomized controlled study, retrospective or prospective comparative studies were included if deemed to contain relevant information. (6) Subgroup analysis in children < 2 years of age undergoing surgery.

Exclusion criteria were: (1) noncomparative studies; (2) studies on patient preparation and preoperative imaging studies as well as articles related to specific clinical situations, such as UPJO in the horseshoe kidney, UPJO in the ectopic kidney, and coexistent urinary lithiasis; (3) pyeloplasty performed on adult or animal; (4) laparoendoscopic single-site surgery and mini-pyeloplasty, salvage procedures such as ureterocalicostomy or ileal ureter; (5) Multiple studies from the same center with data duplication; (6) Case reports, animal experiments, reviews or systematic reviews, meeting minutes and literature with low-quality evaluation.

Types of outcome measures

Primary outcomes

The success rate was defined as stability on postoperative imaging or improved drainage on postoperative diuretic renogram. Intraoperative complications including injury to other organ, blood vessels, and nerves. Postoperative complications including vascular (e.g. haemorrhage, thrombosis), wound (e.g. infection, dehiscence), gastrointestinal (e.g. bowel obstruction due to fibrous adhesions, paralytic ileus due to paralysis of intestinal muscles), incisional hernia, neurological, respiratory (e.g. pneumonia, embolism).

Secondary outcomes

Operating time (from skin incision to closure), the length of hospital stay, and rate of conversion to open surgery.

Data collection and quality assessment

Two evaluators (XHW and KZ) read the title and abstract of the literature, screened the literature according to the inclusion criteria and exclusion criteria, then searched and read the full text, extracted data independently, and cross-checked. The methodological quality of RCTs was assessed by the Cochrane risk of bias tool [13]. The quality of the studies included was determined using the Newcastle–Ottawa scale (NOS) for nonrandomized controlled trials [14]. The Newcastle–Ottawa scale includes selection (4 points), comparability (2 points), and exposure (3 points); They were divided into low risk (7–9 points), medium risk (4–6 points), and high risk (1–3 points), and those with NOS score < 4 points were excluded. The extraction contents included: (1) Basic information of the study: first author, publication year, study area, article source, sample size, age, gender, lesion side, and follow-up time of the children; (2) Outcome measure of the study: success rate, complication rate, operative time, rate of conversion to open surgery, duration of hospital stay.

Statistical analysis

We used the Review Manager v.5.4 software (Cochrane Collaboration, Oxford, UK) statistical package for meta-analysis. The RP group was considered as an experimental intervention, while the LP group was considered a controlled study, 95% confidence interval (CI) represented the effect size, and the P value < 0.05 was considered to be statistically significant. For continuous results, the mean and standard deviation (SD) were used when reported. For primary studies reporting only the median value, the estimated mean of the sample from Luo et al. [15]. The estimated standard deviation of the sample from Wan et al. [16]. Continuous and dichotomous variables were analyzed by weighted mean difference (WMD) and odds ratio (OR), respectively. The heterogeneity was assessed using the Q test (I2 > 75% indicates high heterogeneity, P < 0.1 indicates significant heterogeneity) If no heterogeneity is detected or the heterogeneity is low, the fixed effect model is used; otherwise, the random effect model is used [17]. Explore the source of heterogeneity through subgroup analysis. In addition, sensitivity analysis was carried out on the outcome with high heterogeneity, and the method of eliminating the literature one by one was adopted. After removing the single literature, observe whether there is a significant change in the effect scale and heterogeneity of the remaining research. We used the Mantel Haenszel method to pool dichotomous data and the inverse variance method for continuous outcomes. In this meta-analysis, Review Manager 5.4 was used to draw funnel plot to assess publication bias. If there is no publication bias, the data obtained from each study will show an inverted funnel shape with a symmetric distribution on the graph. In contrast, asymmetric inverted funnel plots show the presence of sample bias. Statistical significance was defined as a P < 0.05.

Results

Study selection

According to the search strategy, 773 articles were obtained, and 18 studies were finally included after eliminating duplicate and irrelevant articles [12, 18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34]. Eighteen studies including 3330 cases (2319 cases for RP and 1011 cases for LP) fulfilled the predefined inclusion criteria. The literature search process and results are shown in Fig. 1. The basic characteristics of the literature and cases included in the meta-analysis are shown in Table 3. Among the 18 included articles, one was a randomized controlled study and 17 were observational cohort studies.

Fig. 1
figure 1

Flow diagram of studies identified, included, and excluded

Full size image
Table 3 The main characteristics of the included studies
Full size table

Success rate

A total of 14 literature recorded the success rate of the two surgical methods. The success rate of RP group was higher than that of LP group, and the difference was statistically significant (OR 2.57, 95%CI (1.24, 5.32), P < 0.01) (Fig. 2). There was no evidence of study heterogeneity (I2 = 0%, P > 0.05).

Fig. 2
figure 2

Forest plot and meta-analysis of success rates

Full size image

Intraoperative complications rate

Five studies compared the incidence of intraoperative complications in the two surgical techniques, among which the most common intraoperative complication was difficulty in double-J tube placement, and the others included bleeding, ectopic vascular injury, and abdominal organ injury. There was no significant difference in the incidence of intraoperative complications between the RP group and the LP group (OR 0.44, 95%CI (0.33, 1.24), P = 0.19) (Fig. 3). The study heterogeneity was potentially irrelevant (I2 = 0%, P = 0.43).

Fig. 3
figure 3

Forest plot and meta-analysis of intraoperative complication rates

Full size image

Postoperative complications rate

There were 14 different studies comparing the two surgical methods' postoperative complication rates, among which the most common postoperative complications were urinary leakage, double J tube displacement, obstruction of double J tube due to impaction of a blood clot, and reoperation required for hydronephrosis aggravation. The overall incidence of postoperative complications of RP was 3.27% (95% CI 2.22–4.31%). The overall incidence of postoperative complications of LP was 6.04% (95% CI 3.5–8.57%). The incidence of postoperative complications in the RP group was lower than that in the LP group (OR 0.61, 95% CI (0.38, 0.99), P = 0.04) (Fig. 4). There was no evidence of study heterogeneity (I2 = 0%, P = 0.98).

Fig. 4
figure 4

Forest plot and meta-analysis of postoperative complication rates

Full size image

Operative time

There are 13 studies comparing the two surgical techniques' operating times. The operation time of the RP group was shorter than that of the LP group, and the difference was statistically significant (MD =  −22.11, 95% CI (− 35.91, − 8.31), P = 0.002) (Fig. 5). The study heterogeneity was considerable and statistically significant (I2 = 81%, P < 0.05).

Fig. 5
figure 5

Forest plot and meta-analysis of operative time

Full size image

Length of hospital stay

A total of 13 articles were comparing the length of hospital stay between the two surgical methods. The postoperative hospital stay in RP group was shorter than that in LP group, and the difference was statistically significant (MD = − 1.04, 95% CI (− 1.6, − 0.47), P < 0.05) (Fig. 6). The study heterogeneity was considerable and statistically significant (I2 = 96%, P < 0.05).

Fig. 6
figure 6

Forest plot and meta-analysis of the length of hospital stay

Full size image

Conversion to open surgery

Only three researchers reported conversion to open surgery. In the RP group, 4 of 219 converted to open surgery, and in the LP group, 6 of 427 converted. Numbers were too small to permit meaningful conclusions (OR 1.09, 95% CI (0.32, 3.79), P = 0.89) (Fig. 7). The study heterogeneity was potentially irrelevant (I2 = 22%, P = 0.28).

Fig. 7
figure 7

Forest plot and meta-analysis of conversion to open surgery rate

Full size image

Subgroup analysis

Subgroup analysis was performed according to patients age. Two studies reported children under 2 years old and sixteen studies reported children older than 2 years old. RP tend to have higher success rate than LP to children older than 2 years old, and the difference was statistically significant. The data of two studies with children under 2 years old showed no significant difference of success rate between two groups (Fig. 8). In children older than 2 years old, RP made fewer postoperative complications than LP, and the difference was statistically significant. The studies with children under 2 years old showed no significant difference of postoperative complications rate between two groups (Fig. 9). In children older than 2 years old, RP are faster than LP at operative time. However, no significant reduction in RP operative time was found compared with LP to children older than 2 years old (Fig. 10). The last outcome of the subgroup analysis showed that the RP group had a shorter hospital stay time than LP group regardless of age (Fig. 11).

Fig. 8
figure 8

Funnel plots illustrating meta-analysis of success rates. SE standard error, OR odds ratio

Full size image
Fig. 9
figure 9

Children under 2 years old showed no significant difference of postoperative complications rate between two groups

Full size image
Fig. 10
figure 10

RP operative time was found compared with LP to children older than 2 years old

Full size image
Fig. 11
figure 11

Subgroup analysis showed that the RP group had a shorter hospital stay time than LP group regardless of age

Full size image

Sensitivity analysis

Sensitivity analysis was conducted for the outcome with large heterogeneity by eliminating the included studies one by one to test the stability of the relevant results. For the length of hospital stay and operation time, the I2 value did not change significantly after removing the studies one by one, indicating that the results were stable.

Publication bias

The success rate was compared between RP and LP, then the funnel plot was generated to detect the publication bias of the included works of literature. The scatter represent of studies were evenly distributed around a straight line, indicating no obvious publication bias (Fig. 12).

Fig. 12
figure 12

Publication bias

Full size image

Discussion

UPJO is the most common congenital malformation of urinary tract in children. UPJO is one of the common causes of hydronephrosis in children, with an incidence of about 1/2000 [35]. With the development of laparoscopic technology, LP has gradually been more and more used in the treatment of UPJO in children [36]. Robot-assisted laparoscopic pyeloplasty was first reported in 2002 after the introduction of the DVSS, the most common robotic procedure in pediatric urology is now RP. This study aims to systematically review and meta-analysis the efficacy of laparoscopic pyeloplasty versus robot-assisted pyeloplasty in the treatment of ureteropelvic junction obstruction in children. Our meta-analysis indicated significantly higher operative success rates, lower postoperative complication rates, and shorter hospital stay as well as operative time in the RP than in the LP group. In 2019, Taktak et al. [37] conducted a meta-analysis of 14 cohort studies, and the results of the meta-analysis suggested that RP had a higher success rate than LP in the pediatric population, but their included literature was small and some of the continuous variable indicators lacked standard deviations and raw data. Our Meta-analysis incorporated newly published comparative studies from recent years and performed subgroup analysis and intraoperative double J-tube placement to demonstrate the advantages of the RP technique in the treatment of children with UPJO. This is an updated meta-analysis of the old reviews concerning the robot-assisted surgery versus laparoscopic surgery of ureteropelvic junction obstruction in children.

For numerous reasons, robot-assisted laparoscopic pyeloplasty has a greater success rate than laparoscopic pyeloplasty in ureteropelvic junction obstruction children over the age of two. To begin, when compared to standard laparoscopic equipment, robotic instruments used in robot-assisted surgery can provide improved accuracy and precision. This is especially critical in pediatric surgery, because the operative region is small and sensitive. Furthermore, robot-assisted laparoscopic surgery can provide improved visibility and magnification, allowing the surgeon to conduct the procedure with minimal injury to adjacent tissues. This can result in improved post-operative outcomes and shorter healing times. Finally, it is possible that this is owing to the enhanced accuracy and precision provided by robotic devices, as well as the use of smaller incisions, which can lower the risk of infection. It is crucial to highlight, however, that each case is unique, and the optimal treatment approach should be determined by the needs and medical history of the individual patient.

Our study showed a higher surgical success rate in the RP group than in the LP group. Ten articles showed that the surgical success rate of the RP group was higher than that of the LP group, and the surgical success rate of both groups was greater than 90% in each study. One article (Franco et al.) indicated that the surgical success rate of RP group was lower than that of LP group, and there was a significant difference in the surgical success rate between RP group and LP group (93.33% vs. 100%) [22]. However, considering that the report time was in 2007, it may be due to the lack of experience of the surgeon and the relative backwardness of robotics and equipment. Although the broad concept of surgical success is clear, that is, there are no clinical or radiographic signs of failure during postoperative follow-up. There are slight differences in the definition of surgical success rate in different research reports. Some authors defined surgical success as the gradual resolution of hydronephrosis or improvement of drainage on renal scans [20]. Others used a renogram only to determine whether the pyeloplasty was successful [30]. One group was based on symptom improvement and the other on radiographic improvement. Despite these differences, we believe that clinical symptoms improvement in children with symptomatic hydronephrosis is a significant marker of successful treatment. With the increasing application of robotic pyeloplasty in children, we need to review and test the efficacy of RP according to existing reports. In 2015, Chang et al. [38] concluded that RP had no obvious advantages over open pyeloplasty. Our meta-analysis indicated significantly higher operative success rates, lower postoperative complication rates, and shorter hospital stay as well as operative time in the RP than in the LP group. In most studies, the success of pyeloplasty is defined as the disappearance of hydronephrosis and symptoms on imaging examination. The problem is that hydronephrosis may last for several months after surgery, and asymptomatic hydronephrosis does not rule out the possibility of obstruction. Although the short-term success rate of RP is higher than that of LP, there is still a lack of long-term follow-up data. The mean duration of follow-up across studies in our meta-analysis was 24 months, which was sufficient to detect early surgical failures. In future publications, the definition of surgical success needs to be standardized, which in turn could lead to an improvement in the quality of the reported success rate. We believe that both RP and LP are effective surgical methods for the treatment of UPJO. With the accumulation of the experience of the surgeons and the continuous development of robotic surgical technology, the surgical success rate of RP is generally higher than that of LP.

The incidence of postoperative complications is an indicator for evaluating the efficacy of surgery and reflecting the prognosis of children. Our study showed that the incidence of postoperative complications in the RP group was lower than that in the LP group. The possible reason may be a clearer field of view during suturing by the robotic operating system, greater suture reliability, and relatively few postoperative complications such as anastomotic fistula. It is also different from Taktak’s study [37], possibly due to the different surgical techniques of individual operators or due to the poor quality of their included literature. Previous studies believe that RP has more advantages than LP in neonatal cases [31]. Because there were only two studies that focused on infants and were included in our meta-analysis. Whether RP has greater advantages in infants needs to be confirmed by studies with larger sample sizes and longer follow-up times. Common complications after RP include urine leakage, displacement of the double J tube, blood clot blocking the double J tube, and reoperation for aggravation of hydronephrosis. Complications of LP include postoperative urinary tract infection, stent or nephrostomy tube displacement, intestinal obstruction, hematuria, and internal organ damage. Due to the overall lower postoperative complication rate of RP, the robotic surgical system may have certain advantages in the treatment of pediatric UPJO.

Operative time is an important parameter when comparing different techniques. Most studies comparing robotic surgery with laparoscopic surgery have assessed surgical team proficiency and surgical skills based on operative time. Our study showed that the operative time of the RP group was significantly shorter than that of the LP group, which was consistent with the results of most domestic and foreign studies. It should be noted that there are differences in the definition of operation time in different kinds of literature, and the main difference is whether to include the installation, debugging, and robot docking time of the robot system. Esposito et al. argued that the robot docking time should be included in the total operative time, as this directly affects the overall anesthesia time of pediatric patients [18]. However, a study found that the operation time of the RP group was longer than that of the LP group, which may be due to the fact that the Da Vinci robot does not have a mechanical feedback system, and the grasping force of the instruments is mainly based on the vision and experience of surgeon [39]. With the continuous accumulation of surgical experience, the continuous improvement of a learning curve [40], and the continuous progress of robotic technology, the actual operation time of RP will continue to be reduced.

Many factors contribute to differences in length of stay, including preoperative preparation, postoperative management, different surgeon preferences, and hospital management. Our meta-analysis showed that the postoperative hospital stay in the RP group was shorter than that in the LP group. On the one hand, the postoperative recovery in the RP group was faster than that in the LP group, and on the other hand, complications might affect the postoperative hospital stay. At the same time, there are differences in medical care programs in different regions, which may also affect the length of hospital stay. From the overall results, children who received RP had faster postoperative recovery and shorter hospital stay. However, there are still few related studies in this area, and further in-depth exploration and analysis are needed.

There are several limitations in our study. There is only one randomized controlled trial among the included studies, which may lead to some bias in the results of our study. The heterogeneity of some outcome indicators is high, which may be due to the inconsistent definitions of outcome indicators in each study, different robot models, and different diseases of patients. Due to the lack of valid data and relevant descriptions in the included literature, the surgical methods could not be subdivided. In the future, researchers can further explore surgical methods in order to obtain more detailed evidence. RP in infants is under development, and it may be too early to compare RP with LP now. As there is a trend of more and more RP performed in infants, we need to review and examine the efficacy of RP based on the available reports.

Conclusion

In conclusion, compared with LP, RP has a higher surgical success rate, fewer postoperative complications, and shorter operative time and length of hospital stay in children with UPJO. However, there were still potential biases in this study. In the future, more randomized controlled trials and long-term follow-up of patients are needed to obtain more reliable analysis results.