Abstract
INTRODUCTION: Point-of-care ultrasonography (POCUS) has emerged as an important tool for examining critically ill patients. POCUS devices have become progressively smaller and more accessible, transforming medical practice, and reducing costs. One technological breakthrough was the development of ultraportable scanners with microchip technology, which utilize a probe connected to a smartphone or tablet and incorporate a mobile application that employs artificial intelligence to assist in the interpretation of acquired images. The aim of this study was to compare the accuracy of ultrasound microchip technology with traditional piezoelectric crystal ultrasonography. We analyzed 2 volume status parameters in intensive care unit (ICU) patients with acute kidney injury (AKI) who were receiving kidney replacement therapy (KRT). These parameters were the extravascular pulmonary water, using the lung B-lines; and the inferior vena cava collapsibility index (IVCi). METHODS: Fifty critically ill patients met the study criteria. Lung POCUS quantified B-lines in 8 quadrants. The IVCi was measured using the maximum and minimum diameters during a respiratory cycle. Both technologies were sequentially employed in a randomized fashion to acquire the parameters at 2 different time points: before the initiation of KRT (T0) and 60 minutes after the procedure had commenced (T60). We calculated the correlation and agreement between the 2 ultrasound scanner modalities. RESULTS: The correlation between the 2 technologies for evaluating lung B-lines showed strong positive coefficients, ⍴ = 0.96 and ⍴ = 0.93 at T0 and T60, respectively (P < 0.001 for both). The correlation for IVCi was ρ = 0.70 and ρ = 0.87 at T0 and T60, respectively (P < 0.001 for both). The Bland-Altman plots showed agreement between ultrasound methods for IVCi calculation and B-line quantification. For IVCi calculation at T0, bias was +2.69 (SD: 10.6) (95% confidence interval [CI]: -18.13 to +23.52); at T60, bias was 3.28 (SD: 10.23) (95% CI: -16.77 to +23.34). For B-line quantification, the analysis yielded a bias of -0.3 (SD: 2.73), with a 95% CI of -5.66 to +5.06 at T0; and a bias of 0.2 (SD: 3.23), with a 95% CI of -6.14 to +6.54 at T60. CONCLUSION: Our study observed a good correlation and agreement between microchip and piezoelectric-based ultrasound modalities in evaluating the presence of pulmonary B-lines and inferior vena cava (IVC) dynamics in patients with AKI. Microchip ultrasound, in addition to being portable, user-friendly, and cost-effective, consistently delivers bedside parameters for volume assessment that are comparable to those obtained through POCUS using piezoelectric transducer-based ultrasonography.