Nasogastric tube placement under sonographic observation: A comparison study of ultrasound and chest radiography in mechanically ventilated patients

Abstract

Background: Nasogastric tube insertion in the intensive care setting is common. Placement verification is required to avoid complications of bronchotracheal misplacement that range from aspiration of infused contents to death from associated causes. The gold standard of practice is chest radiography. Ultrasound is a growing modality and is readily available in most intensive care units. Objective: The objective of this study was to examine the diagnostic accuracy of ultrasound imaging of nasogastric tube placements by nonradiologists compared with chest radiography in mechanically ventilated patients. Methods: This is a dual-centre prospective, single-blind study. Correct placement was captured with a hyperechoic ultrasound image of a nasogastric tube in the oesophagus and epigastrium, which was compared with chest radiography. Patient enrolment included general adult intensive care unit admissions who were mechanically ventilated and required a nasogastric tube for either the treatment or monitoring of their illness. Results: A total of 25 patients were enrolled (15 men, 10 women), and their mean age was 68.1 ± 13.8 years. Outcome measures were the percentage of correctly identified nasogastric tubes in the oesophagus and epigastrium. The sensitivity of oesophagus ultrasound was 88%, and the positive predictive value was 100%. The subxiphoid sensitivity was 64%, and the positive predictive value was 100%. Comparison sensitivity and specificity of oesophagus versus subxiphoid ultrasound was 64% and 33%, respectively. There was a positive predictive value of 88% and a negative predictive value of 11%. The results showed a variance in detection sensitivity in the ultrasound scans of the oesophagus (0.88) and subxiphoid (0.64) (N = 25, p = 0.012). Conclusion: Nasogastric tube placement verification via ultrasound in critically ill mechanically ventilated patients conducted by nonradiologists with minimal training is associated with diagnostic accuracy. These results add to the limited evidence in the current literature; however, they should be considered with awareness that placement in the stomach in this study was detected in 64% of cases, alongside the missed captured evidence of the sonographer's ability to identify misplacement.