Safety of Laparoscopic Gastrostomy in Children Receiving Peritoneal Dialysis.

BACKGROUND

The preferred method of dialysis for children is chronic peritoneal dialysis (CPD), and these children may require delayed gastrostomy tube (GT) placement. Investigators have reported a high risk of fungal peritonitis, early bacterial peritonitis, and catheter loss when percutaneous endoscopic gastrostomy is performed in children already undergoing CPD. Current International Society for Peritoneal Dialysis guidelines recommend only open GT for these patients. We sought to report the safety of laparoscopic gastrostomy (LG) among children already receiving PD.

MATERIALS AND METHODS

We conducted a retrospective chart review of children who had initiated CPD before GT placement between 2010 and 2017 at our pediatric hospital. Demographic data, clinical details, and peritonitis rates were recorded. Peritonitis was defined as peritoneal WBC count >100/mm and >50% neutrophils, with or without a positive peritoneal culture.

RESULTS

Twenty-three subjects had both undergone CPD and had a GT placed in the study period. Of these, 13 had a GT placed after CPD had been initiated. One of these was excluded for open technique and another excluded because of no overlap of GT and PD catheter, leaving 11 for analysis. Median age at the time of LG was 1.32 y and median weight-for-age z-score was -1.86 (IQR -2.9, -1.3). Median days to PD catheter and GT use after LG were 2 (range 0-4) and 1 (range 0-2). Median weight z-score change at 90 d was +0.5 (IQR -0.1, +0.9). All patients received antifungal and antibiotic coverage at time of GT placement. No subjects developed fungal peritonitis or early bacterial peritonitis, although one developed bacterial peritonitis within 30 d. The overall rate of peritonitis after laparoscopic gastrostomy tube was 0.35 episodes/patient-year. This was similar to a rate of 0.45 episodes/patient-year during PD but before laparoscopic gastrostomy tube in the same patients (P = 0.679). Four subjects required periods of hemodialysis, two of which were because of PD catheter removal due to infection. One of the latter was due to a relapse of pre-LG peritonitis and the patient later resumed PD. The other was due to remote post-LG peritonitis and the patient continued hemodialysis until renal transplant, both after 6 mo.

CONCLUSIONS

We found that, in children already receiving PD, LG is similar in safety profile, efficacy, and technical principle to open gastrostomy. LG is therefore an appropriate and safe alternative to open gastrostomy in this setting.