Gastrostomy can be a valuable adjunct to patient care, and percutaneous endoscopic gastrostomy is often considered the method of choice for gastrostomy placement. As with all surgical procedures, however, patient selection is important no matter how the gastrostomy is placed. In a retrospective review of 125 randomly selected patients having gastrostomy tube placement, there were certain groups of patients who received virtually no benefit from gastrostomy and may even have died sooner due to gastrostomy placement. The leading indication for gastrostomy placement was neurologic debilitation; the procedural mortality rate for these patients was 28%. However, patients with pulmonary cachexia or metastatic cachexia had much higher mortality rates: 90% and 37%, respectively. We believe patient selection has been imperfect and that certain patients should not have a gastrostomy tube. These patients suffer the moral indignation of persistent intervention and often die without receiving any real benefit.