A systematic review and meta-analysis of conventional laparoscopic versus robot-assisted laparoscopic pyeloplasty in infants.

OBJECTIVE

While there are several reports confirming the safety and efficacy of laparoscopic pyeloplasty (LP) and robot-assisted laparoscopic pyeloplasty (RALP) in children there have been none comparing LP and RALP specifically in infants. In this meta-analysis, we have compared the outcomes of LP and RALP in infants.

METHODS

Pubmed (Medline), Publon, Index Medicus and Embase were searched using the search terms: pyeloplasty (laparoscopic OR robot-assisted) AND (infant), to identify all papers pertaining to LP and RALP. Systematic review was performed to identify information regarding number of patients/renal units, age, body weight, operating time, hospital stay, success and complications. Meta-analysis of heterogeneity was reported with I2statistics. Once heterogeneity was found low, the pooled outcomes were compared with student's t test and Fishers exact test, wherever appropriate.

RESULTS

After screening a total of 267 articles, 18 articles were included (10 articles on LP, 7 on RALP, 1 reporting both), comprising 323 renal units for LP and 173 renal units for RALP. With low heterogeneity (I2: 0%) both groups were considered to have been conducted under similar conditions for fixed effect model. There was no significant difference between the success rates of LP or RALP (97.5% vs 94.8%; p = 0.21). The mean age at operation was significantly lower for LP (5.6 ± 1.8 months) than RALP (7.2 ± 1.2 months, P = 0.0001). The duration of surgery was 137 ± 45 min for LP while significantly higher at 179 ± 49 min for RALP (p = 0.0001). The mean (s.d) time to discharge was 2.0 (1.9) days for LP while 1.3 (0.4) days for RALP. The overall complication rate was significantly higher (summary table) for RALP than LP (p = 0.03), mainly due to more port-site hernias in RALP.

DISCUSSION

In the present study, we found that the success of LP and RALP in infants was similar. RALP in infants had longer duration of surgery, similar hospital stay and higher Clavien-3 complications than LP. While several studies have reported favorable outcomes for RALP over LP in children, this was not the case in infants. The smaller workspace, in an infant, can significantly limit the mobility of robotic instruments and increase the chance of port-site conflicts or trocar collisions. The use of larger robotic ports and instruments in the small space of infant abdomen might have been responsible for higher complications in RALP, including significantly larger number of port-site hernias. This meta-analysis represents the early experience of most RALP in infants, and it is possible that with experience RALP outcomes in infants also will catch up with LP. Miniaturization of robotic instruments might render RALP the future standard of care for pyeloplasty in infants.