Pediatric laparoscopic versus percutaneous gastrostomy tube placement: a single-center review

Background: The optimal technique for gastrostomy tube (GT) placement in pediatric patients remains controversial. Percutaneous endoscopic gastrostomy (PEG) was the preferred approach over open gastrostomy. With the advent of laparoscopy, many advocate for laparoscopic (LAP) placement to avoid potential visceral injury. Additionally, PEG patients may undergo an additional procedure for conversion to a low-profile button. We sought to compare outcomes including complications, need for subsequent procedures, and anesthesia exposure in LAP vs. PEG patients.

Methods: Patients (ages 0-18) who underwent GT placement at our pediatric healthcare system between 2018 and 2021 were retrospectively reviewed. Patients were excluded if they underwent fundoplication, gastro-jejunostomy tube placement, open placement, tube placement in concurrence with other intestinal procedures, or failed primary attempt at gastrostomy placement. Data related to demographics and GT placement were recorded. Our primary outcomes were complications, need for subsequent procedures, discrete anesthesia exposures, and cumulative anesthesia exposure. The Wilcoxon rank sum test, Pearson's Chi-squared test, and Fisher's exact test were used to compare characteristics and clinical measurements between PEG and LAP patients.

Results: Six hundred and eighty-eight (688) patients underwent GT placement during the study period, 234 (34.0%) LAP and 454 (66.0%) PEG. LAP patients were younger and weighed less than PEG patients (p = 0.005 and p = 0.002, respectively). Gender distribution, primary insurance status, and ASA (American Society of Anesthesiologists) classification were similar. Within the group excluded, 5 failed PEG placements, while 0 failed LAP GT attempts (p = 0.173). Major complication rates were comparable (1.3% vs. 2.4%, p = 0.401); however, PEG patients were more likely to have skin erythema/local infection (p = 0.006). PEG patients tended toward undergoing subsequent procedures (10.9% vs. 6.5% for LAP, p = 0.061) such as GT revision or conversion to gastro-jejunostomy tube. Additionally, 60.5% of PEG patients required > 2 anesthesia events, most often due to exchange of PEG to a low-profile button, while 93.6% of LAP patients required only one (p < 0.001). Finally, the median total general anesthesia exposure for the PEG group was 75 min (IQR 53-97) and 79 (IQR 67-98) in the LAP group (p = 0.002).

Conclusion: PEG technique is associated with more discrete anesthesia exposures and may also require more subsequent operations related to its placement. However, at our institution, overall major complications are similar in both techniques, while PEG tubes are prone to skin erythema/local infection.

Level of evidence: Retrospective Comparative Study, Level III.