Purpose: Gastrointestinal obstruction presents many burdens for patients with end-stage abdominal cancer, such as nausea and vomiting. Few detailed data on the efficacy of a percutaneous endoscopic gastrostomy (PEG) for decompression exists. This retrospective cohort study investigates the quantity of symptom relief realized with PEG and the corresponding complications. Methods: Chart reviews of 75 patients with malignant gastrointestinal obstruction, who received a PEG for decompression, were performed. Abstracted data includes symptoms (vomiting, nausea, abdominal pain) and medication up to 7 days before and after the intervention, complications, demographics, potential influencing factors and survival. Generalized estimating equations (GEE) models determined symptom reduction. Results: PEG decreased the mean frequency of vomiting per day from 2.2 (95% confidence interval (CI) 1.7–2.7) to 0.4 (95% CI 0.3–0.6) (p < 0.001). The probability of the occurrence of nausea on a given day was 80% (95% CI 74–85%) prior to the PEG placement and 40% (95% CI 34–47%) afterwards (p < 0.001). One hundred twelve complications were reported in 56 patients (none 19/75 patients (25%), minor 52/75 (69%), major 18/75 (24%)). Stomal leakage (18/75 patients), mild wound pain (17/75) and tube occlusion (13/75) occurred most frequently. The failure of the first attempt of the PEG placement (7/75) presented as the leading major complication. Conclusions: The PEG for decompression significantly reduces vomiting and nausea in patients with malignant gastrointestinal obstruction (p < 0.001). Minor complications are common and should be discussed prior to the intervention. Nevertheless, the PEG appears to demonstrate prevailing benefits in comparison to the risks.