Feasibility of Robotic Surgery in Comparison to the conventional Laparoscopic and Open Procedures in Urology: A Systematic Review

Afra Zahid King Edward Medical College: King Edward Medical University Muhammad Ayyan (  iamayyan77@gmail.com ) King Edward Medical College: King Edward Medical University https://orcid.org/0000-0002-4023-7956 Minaam Farooq King Edward Medical College: King Edward Medical University Huzaifa Ahmad Cheema King Edward Medical College: King Edward Medical University Abia Shahid King Edward Medical College: King Edward Medical University Faiza Naeem King Edward Medical College: King Edward Medical University Muhammad Abdullah Ilyas King Edward Medical College: King Edward Medical University Shehreen Sohail University of Central Punjab


Introduction
The use of Arti cial Intelligence (AI) in the eld of surgery has been increasing over the past few decades. The rst robot ever used for surgery was the Unimation PUMA 200 to assist a neurosurgical intervention performed by Kwoh et al. in 1985[1] while the rst urological procedure for which a robot was used was transurethral prostate resection, performed by the assistance of PROBOT in 1989 [2,3]. AESOP (Automated Endoscopic System for Optimal Positioning) was the rst surgical robotic system that became FDA approved in 1993 and commercially available for use in surgical procedures, including those in the eld of of the age. 6) The articles including patients belonging to any age group who underwent surgical interventions for urological conditions. 7) The articles on the feasibility of robotic surgery in the eld of urology in terms of the outcome measures (primary, secondary and additional).
The studies were excluded according to the following criteria: 1) Systematic reviews, meta-analysis and other study designs except Randomized Controlled Trials (RCTs) and Cohort Studies. 2) Unpublished articles and those published in languages other than English. 3) Studies conducted on non-human populations. 4) The studies not dealing with robotic interventions. 5) The articles that did not deal with urological procedures. 6) The articles that did not deal with the feasibility of robotic surgery in terms of the outcome measures (primary, secondary and additional). 7

SEARCH STRATEGY
The database search was done using the terms "Robotic surgical procedures", "Robotic surgery", "Urology" "Robot Assisted Prostatectomy", "Robot Assisted Cystectomy" etc. and their combinations.
The detailed search strategy is presented in Table 1.

STUDY SELECTION
After the database search, the articles were imported to EndNote X9 Software and de-duplication was done.
Two independent reviewers screened the articles, initially according to the title and abstract and later based on the full text. The articles were selected based on the inclusion criteria. Any con ict during the screening process was resolved by discussion. In case of unresolved disagreements, it was planned that the third author would make the decision.

PRIMARY OUTCOMES
The primary outcomes were: (i) oncological outcomes (For example: positive surgical margins, recurrence rate and mean lymph node yield); (ii) functional outcomes (For example: time to atus, renal function (reduction of GFR), urinary and sexual function); and (iii) complications.

SECONDARY OUTCOMES
The secondary outcome measures comprised of: (i) mortality; (ii) length of hospital stay; (iii) blood loss; (iv) mean operative time; (v) warm ischemia time; and (vi) need for postoperative analgesia.
The additional outcome was cost effectiveness.

For Cohort Studies
The Newcastle-Ottawa Scale (NOS) checklist [17] was used. It consists of three categories, namely Selection, Comparability and Representativeness, on the basis of which the cohort studies were divided into three levels of scoring: poor quality (1 to 4 stars), fair quality (5 to 7 stars) and good quality (8 to 9 stars).

For Randomized Controlled Trials (RCTs)
The quality assessment of the Randomized Controlled Trials was performed using the revised Cochrane "Risk of Bias" tool for randomized trials (RoB 2.0)[18]. RoB 2.0 consists of ve speci c domains: (1) bias arising from the randomization process; (2) bias due to deviations from the intended interventions; (3) bias due to missing outcome data; (4) bias in measurement of the outcome; and (5) bias in the selection of the reported result.
Using RoB 2.0, each speci c outcome was assessed and its overall Risk of Bias was found (low; some concerns; high). The overall Risk of Bias was derived using the highest RoB levels in any of the domains that were tested.
The quality assessment of each study was performed by two review authors independently. Any disagreements were resolved by discussion. In case of unresolved disagreements, a third review author was consulted to make the decision.

DATA SYNTHESIS
A qualitative synthesis of the data extracted from the included studies was performed. The data was presented in a tabulated form and included study variables, outcome variables, quality assessment and the conclusions of each individual study. Meta-analyses could not be conducted because of the heterogeneity of interventions, participants and outcome measures.

STUDY SELECTION
A total of 12,088 records were found through databases searching. After de-duplication, 10 [51,52]. The study characteristics are summarized in Table 2. assessed to have a good quality (low risk of bias) while 2 (6.5%) [31,41] were assessed to have a fair quality (medium risk of bias). The most common methodological aws were inadequate representativeness of the exposed cohort, non-adjustment for the confounders and inadequate duration of the follow-up. The ndings are summarized in Table 3.
The quality assessment of the 3 Randomized Controlled Trials (RCTs) was performed using the revised Cochrane "Risk of Bias" tool for randomized trials (RoB 2.0). The overall risk of bias was assessed to be "low" for 1 (33.3%) study [27] and "some concerns" for the other 2 (66.6%) [24,26]. The most common methodological aws were lack of blinding of participants and outcome assessors, improper allocation concealment, deviations from the intended interventions and lack of information regarding pre-speci ed analysis plan. The RoB plot was created using the Robvis tool [53]. [ Figure 2]

RESULTS OF INDIVIDUAL STUDIES
The results of each individual study included in the review are presented in Table 4. Robot-assisted prostatectomy was found to take a longer time (2.3 hours for conventional radical retropubic prostatectomy versus 4.8 hours for robot-assisted prostatectomy (p < 0.001)) but associated with less blood loss (970 mL for RRP versus 329 mL for RAP (p < 0.001)), early discharge from the hospital and less postoperative pain (mean pain score on postoperative day 1 was 7 in the RRP group and 4 in the RAP group (p = 0.05). The complications and oncological outcomes were found to be similar to those of radical retropubic prostatectomy.

Jong Wook Park et al, 2011
Both robot-assisted and laparoscopic prostatectomy had similar safety and e cacy but the robotic approach was associated with faster recovery of continence, although the continence rate at 12 months became similar (95% for LRP versus 94.4% for RALP, p = 1.00).

Guillaume Ploussard et al, 2012
The robot-assisted approach was found to be associated with shorter operative time (129 versus 175 min; p < 0.001), less blood loss (515 versus 800 ml; p < 0.001), and shorter hospital stays (4.0 versus 5.7 days; p < 0.001) as compared to the pure laparoscopic approach. Continence was unaffected by the type of procedure performed while there was an early recovery of potency by the robot-assisted approach (After 6 months, 20% patients were potent after LRP and 42% after RALP. After 12 months, 31.6% of patients were potent after LRP and 57.7% after RALP)

Fransesco Porpiglia et al, 2016
Robot-assisted prostatectomy was associated with better functional outcomes in terms of recovery of continence (p < 0.021) and potency (p < 0.028) but was found to be similar to the laparoscopic approach in terms of the oncological outcomes and major complications. The robotic approach was associated with some perioperative bene ts as compared to the laparoscopic approach. RARC group had lesser operative time (p < 0.001), lesser blood loss (p < 0.001), lower intraoperative transfusion rate (p < 0.05), shorter hospital stays (p < 0.001), and a lower 90-day complication rate compared with the LRC group, but there was no difference in the rate of readmissions between the two approaches.

Janet Baack Kukreja et al, 2020
The robot-assisted approach was more expensive as compared to the open approach but was associated with lesser costs per quality-adjusted life years (RARC was $17 000 more expensive, but associated with an increase of 0.32 QALYs)

Tanya Nazemi et al, 2006
The robot-assisted approach was said to be associated with lower blood loss (p = 0.01) and analgesia use (open nephrectomy compared to robotic and hand-assisted laparoscopic methods at 75% versus 0% and 14%, for the robotic and hand-assisted laparoscopic methods respectively (p = 0.0035)) and shorter hospital stay (3 vs The robot-assisted approach was similar to the laparoscopic approach in terms of the oncological and perioperative complications but was associated with a slightly longer operative time (221 versus 175.3 minutes in robot-assisted and laparoscopic methods, respectively, p = 0.001).

In Gab Jeong et al, 2017
The robot-assisted approach was associated with higher costs as compared to the other approaches for nephrectomy ( Robot-assisted partial nephrectomy was associated with the advantages of threedimensional vision and easier suturing over the traditional laparoscopic approach. The mean warm ischemia time was found to be signi cantly shorter in the RAPN group (27.3 minutes for the RAPN group and 35.8 for the LPN group) (p = 0.02). Some demerits included higher costs and the need for surgeons with experience in the eld.

Brian M Benway et al, 2009
Robot-assisted partial nephrectomy provided comparable pathological outcomes and morbidity as compared to traditional laparoscopy. However, it was associated with reduced blood loss (155 for RAPN versus 196 ml for LPN, p = 0.03), warm ischemia time (19.7 versus 28.4 minutes, p < 0.0001), and length of hospital stay (2.4 versus 2.7 days, p < 0.0001). It was also found to be supposedly superior to the laparoscopic approach while dealing with more complex tumors.

Agnes J Wang et al, 2009
The robot-assisted procedure was comparable to the laparoscopic partial nephrectomy in terms of postoperative outcomes but was associated with shortened warm ischemia time (19 versus 25 minutes, p = 0.03). Some disadvantages included higher cost, use of extra trocars (4.6 versus 3.2, p = 0.01), and the more invasive nature of the robotic approach.

Sangchul Lee et al, 2011
Both the robotic and open approaches were associated with similar changes in GFR (83.3 for RPN versus 79.6 mL/min/1.73 m 2 for OPN, p = 0.146) but the robotic surgery was preferred due to shorter length of hospital stay (6.2 versus 8.9 days, p < 0.001) and lesser analgesic use Robot-assisted partial nephrectomy was found to be associated with the advantages of having shorter operative (152 versus 193 minutes, p < 0.001) and ischemia times (14.0 versus 18.0 minutes, p < 0.001) as well as reduced blood loss (122 versus 245 mL, p = 0.001) as compared to the traditional laparoscopic approach.

Ali Khalifeh et al, 2012
The robot-assisted approach was found to have the advantages of having lower intraoperative complications (2.6% for the robotic approach versus 5.6% for the laparoscopic approach, each p < 0.001), lower postoperative complications (24.53% for the robotic approach versus 32.03% for the laparoscopic approach, p = 0.004) and lower risk of positive surgical margins (2.9% for the robotic procedure versus 5.6% for the laparoscopic procedure, p < 0.001) as compared to the conventional laparoscopic method of surgery.

Vincent Trudeau et al, 2014
The robot-assisted approach was found to be more costly ($23,235 for the robotic approach versus $17,637 for the laparoscopic approach; p < 0.001) but associated with lesser complications (11.9% in the robotic approach versus 18.2% in the laparoscopic approach; p < 0.001) as compared to the traditional laparoscopic approach.

Hakmin Lee et al, 2019
The laparoscopic and robotic showed some better perioperative outcomes (all p-values > 0.05) after radical nephroureterectomy as compared to the open approach in patients with non-metastatic UTUC. The recurrence rate was comparable in all the groups (p = 0.279). Robot-assisted adrenalectomy was found to be a feasible approach since it was associated with lesser use of analgesics (43 mg in Robotic Adrenalectomy versus 84 mg in Laparoscopic Adrenalectomy group, p < 0.001) but comparable costs (84% ± 14% in Robotic Adrenalectomy versus 100% ± 16% in Laparoscopic Adrenalectomy)

Kai Alexander Probst et al, 2016
The robot-assisted approach was found to be feasible for the dissection of adrenal tumors. Although it was associated with longer operative time (128.5 ± 46.5 min for RALA versus 102.2 ± 44.5 min for OA), the overall hospital stay was found to be shorter for the robotic approach (6.8 ± 1.2 days for RALA versus 11.1 ± 4.8 days for OA).

Amitabh Singh et al, 2018
The robot-assisted approach was found to be associated with lower morbidity and comparable lymph node yield (13 in Robot-assisted procedure versus 12.5 in open inguinal lymph node dissection) as compared to the open approach.

Hualiang Yu et al, 2019
The robot-assisted approach was found to be associated with fewer intraoperative and postoperative complications while achieving the desired outcomes.

RESULTS OF SYNTHESES
A qualitative narrative synthesis of the data extracted from the included studies was performed and presented in the form of tables. There was not a scope for a meta-analysis, given the likely range of interventions, participants and outcomes. The results for the feasibility of each surgical intervention according to the outcomes have been presented in Table 5. Two studies reported faster recovery of continence in the robotic approach as compared to the other approaches while one reported greater urinary incontinence for the robotic procedure.
Recovery of sexual function

(Three Studies)
Two studies reported faster recovery of potency in the robotic approach while one reported greater erectile dysfunction for the robotic approach.

Complications 11389 (Four Studies)
Two studies reported lesser complications for the robotic approach while one reported no difference. According to one study, the rate of genitourinary complications was higher in the robotic approach.
Postoperative analgesia 8837 (One Study) Reduced need for postoperative analgesia as compared to the other approaches. According to one study, the lymph node yield was comparable in the robotic and other approaches while one study reported greater lymph node yield as compared to the laparoscopic approach but lesser as compared to the open approach.
Recurrence Rate

(One Study)
Greater as compared to the other approaches.
Mortality 60 (One Study) According to one study, the mortality for the robotic-assisted procedure was found to be higher than the open approach but lower than the laparoscopic approach. Comparable to the other approaches.
Complications 8316 (One Study) Fewer perioperative complications for the robotic approach.
Postoperative analgesia

(Two Studies)
According to one study, the need for postoperative analgesia was less as compared to the conventional methods while one study reported comparable postoperative analgesia use.

Estimated
Blood Loss

(Two Studies)
According to one study, the estimated blood loss was less as compared to the conventional methods while one study reported comparable blood loss.

Mean
Operative Time

(Two Studies)
Greater as compared to the conventional methods.

(Two Studies)
According to one study, the length of stay was shorter as compared to the conventional methods while one study reported a comparable length of stay. According to one study, the decrease in GFR was less as compared to the conventional methods while one study reported comparable changes in GFR.

Complications 998 (Five Studies)
Two studies reported decreased complications as compared to the other methods while two reported one complication each for the robotic approach. According to one study, the robotic approach was associated with increased complications, the most common being urinary leakage while one reported comparable intraoperative complications.
Postoperative Analgesia 303 (One Study) Reduced need for postoperative analgesia.

Estimated
Blood Loss

(Six Studies)
Three studies reported reduced blood loss for the robotic approach while according to three studies, the blood loss was similar for all approaches.

Mean
Operative Time

(Seven Studies)
According to four studies, the operative time was shorter for the robotic approach while the others reported comparable, or even increased operative time.

(Three Studies)
The length of stay was found to be shorter for the robotic method by two studies but one reported similar results for the different approaches.

Warm
Ischemia Time

(Seven Studies)
According to ve studies, the warm ischemia time was shorter for the robotic approach while the others reported comparable, or even increased warm ischemia time. Shorter as compared to the conventional methods.

Discussion
The use of robots for surgical procedures has been gaining popularity, especially in certain elds including urology. This has become an important topic for researchers, and many studies have reported their superiority as compared to the conventional open and laparoscopic methods. Our study explains the differences between the robotic and other approaches in terms of outcome measures comprehensively, covering all the urological procedures. Our results are consistent with the fact that robot assisted surgeries, being minimally invasive, have the advantage of being associated with fewer complications, lesser blood loss, lesser need of postoperative analgesia and shorter length of hospital stay. The mean operative time was found to be longer for robot assisted surgeries due to extra time required for robot speci c tasks, such as set-up of equipment and longer trocar entry time [66]. For some of the outcome measures, our results differed from those of other systematic reviews and meta-analyses because of the differences in the study settings and the surgeon skill set of the studies included. The differences can also be attributed to a bias in the selection of the studies that were included in the reviews.
Most of the studies included in our review were limited by short periods of follow up and lack of randomization. Other limitations included small sample size and the retrospective nature of some studies, which introduced an inherent selection bias. Some of the included cohort studies were not controlled for confounders.
Our review was registered in the International Prospective Register of Systematic Reviews (PROSPERO). A highly sensitive search strategy was applied for the selection of articles and all the robotic surgical procedures in the eld of urology have been included. There were certain limitations of the review process. A meta-analysis could not be conducted owing to the signi cant heterogeneity between the study results. Only the articles that were published in English language were included. The number of RCTs included was less, and so was the number of articles of each surgical procedure. The assessment and reporting of outcomes was not well standardized. However, we are con dent that these limitations would not have an effect on the conclusions of the review.
According to our review, the robotic approach can be a good alternative to the conventional procedures due to the advantage of being associated with shorter length of hospital stay, lesser estimated blood loss, reduced need of postoperative analgesia and fewer perioperative and postoperative complications.
Our review can be used by patients, urologists and policy makers to make decisions regarding robotic surgical procedures by weighing the pros and cons and establishing its feasibility. There is an existing research gap when it comes to robotic surgery, especially in procedures like Inguinal Lymphadenectomy, Ureteral Reimplantation, Nephroureterectomy and Adrenalectomy.
More Randomized Controlled Trials (RCTs), especially those dealing with outcome measures like oncological and postoperative measures, and those with longer follow up duration are required to better ascertain the feasibility of robotic surgical procedures.

Conclusion
The feasibility of Robot Assisted surgery varied for different outcome measures as well as for different procedures. Some of the common advantages included shorter length of hospital stay, lesser estimated blood loss, reduced need of postoperative analgesia and fewer perioperative and postoperative complications while the disadvantages included longer operative time. The feasibility of Robot Assisted procedures can be better determined by more and larger clinical trials.

Declarations FUNDING
We declare that no funds or grants were received during the preparation of this review.

COMPETING INTERESTS
We do not have any relevant nancial or non-nancial interests.

AUTHOR CONTRIBUTIONS
Afra Zahid, Muhammad Ayyan, Huzaifa Ahmad Cheema and Minaam Farooq contributed to the conception and design of the review and screening of the articles to be included. Abia Shahid, Faiza Naeem, Abdullah Ilyas and Shehreen Sohail performed the analysis and interpretation of the included studies. All the authors contributed to the drafting, reviewing, and nal approval of the manuscript.

ETHICS APPROVAL
No ethical approval was required for this study.
Abbreviations And Acronyms Table 3   Table 3 is available in the Supplemental Files section

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Table3.docx