A Systematic Review and Meta-Analysis on Outcomes and Complications of Percutaneous Endoscopic Versus Radiologic Gastrostomy for Enteral Feeding.

BACKGROUND

The optimal technique for long-term enteral feeding has not yet been established. Both percutaneous endoscopic gastrostomy (PEG) and percutaneous radiologic gastrostomy (PRG) are widely used. Aim was to extensively review outcomes of PEG and PRG.

MATERIALS AND METHODS

A systematic review using Medline, Embase, and Cochrane was performed, using standardized tools for assessing bias. Main outcomes were infectious and tube-related complications, procedure related and 30-day mortality. Pooled risk differences (RDs) with corresponding 95% confidence intervals (95% CIs) were calculated using random effects. Arcsine transformations were applied.

RESULTS

In total, 344 studies were identified, of which 16 were included, reporting on 934 PEGs and 1093 PRGs. No differences were found for infectious complications [RD, 0.03 (-0.05 to 0.11)], procedure-related mortality [RD, 0.01 (-0.04 to 0.06)], or 30-day mortality [RD, 0.06 (-0.01 to 0.13)]. Tube-related complications were higher in PRG [RD, 0.16 (0.06-0.26)]. Subgroup analysis was performed for head and neck cancer (HNC) and motor neuron disease. In HNC, this revealed significantly lower tube-related complications and procedure-related mortality after PEG. In motor neuron disease, no differences were seen. The level of evidence appears sufficient considering the low degree of heterogeneity.

CONCLUSIONS

No differences were found with regard to mortality or infectious complications. PEG showed lower risk of tube-related complications. Subgroup analysis revealed PEG to be favorable in HNC based on lower rates of procedure-related mortality and tube-related complications. Local experience and availability should be taken into account in the decision process.