It is still unclear, however, if the accuracy of LUS in diagnosing ACD may be higher when performed by more skilled physicians. All the LUS in our study were performed in duplicate (and blinded) by a trained nurse and by a physician expert in LUS; we thus compared the overall accuracy of these two examinations by using as a reference the overall final diagnosis as assessed by two external independent physicians (J.V. and F.D.) through review of the entire medical record of each patient. In the diagnosis of CHF, physician-performed LUS revealed a similar sensitivity (95.3% vs 95.3%) and a nonsignificantly higher specificity (93.3% vs 88.2%), with a resulting nonsignificantly higher area under the curve (0.94 [95% CI, 0.91-0.98] vs 0.92 [95% CI, 0.88-0.96]) compared with nurse-performed LUS (Table 1). Furthermore, there was a good correlation between ultrasonographic results collected by nurses and physicians (Cohen’s κ = 0.73 [95% CI, 0.67-0.79]). Thus, our results further confirm the good accuracy of nurse-performed LUS, suggesting this technique should be used when a trained physician is not readily available and to reduce the time of diagnosis in overcrowded emergency departments. Nevertheless, nurses should have adequate training in this technique before safely and efficiently applying LUS in clinical practice.