Introduction: Diaphragm paralysis or paresis (DP) is a well-studied complication after congenital heart surgery due to the increased risk of injury to the phrenic nerve, with an incidence described in the literature between 0.5% and 13% or after birth trauma in neonates. Injury to the phrenic nerve is hard to detect immediately after the procedure, and is often a diagnosis of exclusion with failure to wean off mechanical ventilation after cardiac surgery. This often leads to increased length of stay in the Intensive Care Unit (ICU), morbidity from hospital stay, and prolonged periods of mechanical ventilation. Treatment involves surgical plication of the affected hemidiaphragm to prevent paradoxical motion and assist the ventilation mechanics, further increasing the length of stay. The gold standard for the diagnosis of DP is video fluoroscopy or electromyography of the affected side. These procedures often need the child to be off positive pressure ventilation and leave the ICU for the testing to be satisfactorily performed. Bedside ultrasonography (US) is a quick and non-invasive tool to detect abnormality of motion of the hemidiaphragm. We sought to review the available literature comparing the performance of the US as a diagnostic tool for the diagnosis of DP. Methods: We examined literature published over the last thirty years and screened 46 studies. We included studies that looked at US application and methodology in assessing diaphragm motion and trials that compared US performance against fluoroscopy. Results: Of the 46 studies, only 7 studies (15%) were robustly designed trials comparing the performance of US vs the gold standard. These studies reported 98-100% sensitivity and variable specificity in predicting fluoroscopic diagnosis of DP. Conclusions: Based on available literature, diaphragm ultrasound, with the inclusion of M-mode, is equally effective, if not superior to, fluoroscopy in early screening and detection of DP. The evidence can be greatly improved by a large multicenter double-blinded trial to establish the specificity of bedside US.
Vijayakumar et al. (Sun,) studied this question.