Diagnostic accuracy of nurse‑performed lung ultrasound for pulmonary congestion in acute kidney injury: An exploratory study.

Abstract

Purpose: To estimate the diagnostic accuracy of nurse-performed lung ultrasound (LUS) for detecting pulmonary congestion in adults with acute kidney injury (AKI) and to relate ultrasound findings to the defining characteristics of the NANDA-I nursing diagnosis excess fluid volume.

Methods: In this prospective exploratory diagnostic accuracy study, conducted between October 2022 and September 2023, a critical care nurse performed bedside LUS, following the bedside lung ultrasound in emergency protocol, in a convenience sample of 64 intensive care unit patients with AKI in a general hospital in Brazil. Pulmonary congestion was defined as ≥3 B-lines in ≥2 intercostal spaces per hemithorax. The reference standard was radiological evidence of vascular congestion on chest radiograph or computed tomography interpreted by blinded intensivists. Sensitivity, specificity, positive and negative predictive values were calculated with 95% confidence intervals. Agreement between LUS and radiologic findings was assessed using Gwet's AC1 coefficient. The COVID-19 pandemic limited patient flow; therefore, no formal sample-size calculation was feasible. Results should be considered preliminary.

Findings: Pulmonary congestion was present in 14/64 patients (21.9%). LUS sensitivity was 50% (95% CI 23%-77%) and specificity 94 % (89%-99%). Positive and negative predictive values were 70% and 87%, respectively. Agreement between LUS and radiology was substantial (AC1 = 0.77, 0.63-0.92). No LUS-related adverse events occurred. Presence of ≥3 B-lines-an objective ultrasonographic marker of interstitial fluid-corresponded to the defining characteristic "pulmonary congestion" of NANDA-I diagnosis 00026.

Conclusions: A single well-trained nurse achieved high specificity using LUS to rule out pulmonary congestion in critically ill patients with AKI. Limited sensitivity and wide confidence intervals highlight the need for larger, multicenter studies with sufficient positive cases. Linking B-lines to the defining characteristic of Excess Fluid Volume supports integration of point-of-care ultrasound into nursing diagnostic reasoning.

Implications for nursing practice: Bedside nurse-performed LUS can strengthen clinical decision-making by quickly excluding pulmonary congestion-thereby informing volume-management interventions aligned with the NANDA-I taxonomy.