Chest X-ray imaging is frequently used for verifying the position of a blindly inserted nasogastric tube. A high-quality X-ray increases the likelihood of conclusive visibility of tube tip positioning, thus avoiding risks due to a misplaced tube (e.g., pulmonary intubation, pneumothorax, small bowel insertion). Therefore, this study aims to determine patient-related and environmental factors affecting the visibility of nasogastric tubes on X-ray in adults. A retrospective descriptive analysis of routinely collected clinical data was performed on all included patients (N = 215) from a prospective randomized trial in a general hospital. A chest X-ray was taken of each patient needing a nasogastric feeding tube, after which visibility and positioning of the tube on X-ray was independently evaluated by 3 radiologists. In 14.9% (n = 32) of all patients, image quality was insufficient, so no conclusive visibility of nasogastric tube positioning could be found. A patient-related predictor regression model (sex, age, body mass index) explained 21% of variance for an insufficient visibility of the nasogastric tube (Nagelkerke R2 = 0.21). An environmental factor regression model demonstrates a guidewire being inside the tube or not during X-ray as a predictor for a conclusive visibility on X-ray. High body mass index, male sex, and the absence of a guidewire inside the nasogastric tube at the time of chest X-ray are associated with a risk of insufficient visibility of the tube on X-ray. Patient profiles can be defined in which supplementary attention is needed when obtaining chest X-rays whose purpose is to confirm nasogastric tube positioning. • The quality of chest X-rays to confirm the positioning of nasogastric tubes in adults can be improved considerably. • There are several factors influencing the confirmation of nasogastric tube positioning on X-ray. • Defining patient profiles at risk for an insufficient visibility of the tube on X-ray will indirectly contribute to an improvement of the chest X-ray quality.
Read full abstract