Tumor implantation following percutaneous endoscopic gastrostomy insertion for head and neck and oesophageal cancer: Review of the literature.

BACKGROUND: Because of publication bias, there is uncertainty about the true incidence of tumor seeding or implantation in patients with head and neck or oesophageal cancer undergoing percutaneous endoscopic gastrostomy (PEG) insertion.

METHODS: In order to obtain a more reliable estimate of risk, a systematic review was undertaken. Randomized or non-randomized studies and case reports were identified by electronic searching. A risk of bias assessment was carried out for each study.

RESULTS: Ninety-eight cases from 74 published case reports and 1 unpublished case were identified. Synchronous distant metastases were present in 37%. Analysis of case series (6192 patients) considered to carry a moderate risk of bias suggests an incidence of seeding after PEG insertion of 0.32%. Studies carrying a lower risk of bias indicate a risk of seeding closer to 1 in 2000.

CONCLUSION: The true risk of seeding after PEG insertion is probably less than 1 in 1000.