Skip to main content

Table 2 Summary of comparisons made between measurement of the inferior vena cava (IVC) in healthy volunteers

From: Assessment of the inferior vena cava collapsibility from subcostal and trans-hepatic imaging using both M-mode or artificial intelligence: a prospective study on healthy volunteers

Comparison

Variable

Mean Bias and 95% CI

LoA and 95% CI

ICC and 95% CI

Lower

Upper

AI-SC

M-SC

IVCc

− 0.7% [− 4.1 to 2.7]

− 24.9% [− 30.8 to v19.1]

23.6% [17.7 to 29.5]

0.57 [0.36 to 0.73]

IVC-max

− 2.5 mm [− 3.1 to − 2.0]

− 6.7 mm [− 7.7 to − 5.7]

1.6 mm [0.6 to 2.6]

0.83 [0.73 to 0.90]

IVC-min

− 1.4 mm [− 2.1 to − 0.7]

− 6.3 mm [− 7.5 to − 5.1]

3.5 mm [2.4 to 4.7]

0.81 [0.69 to 0.89]

AI-TH

M-TH

IVCc

3.7% [1.0 to 6.3]

− 14.9% [− 19.5 to − 10.3]

22.3% [17.7 to 26.8]

0.72 [0.55 to 0.83]

IVC-max

− 2.4 mm [− 3.2 to − 1.5]

− 8.1 mm [− 9.5 to − 6.7]

3.4 mm [1.9 to 4.8]

0.67 [0.49 to 0.80]

IVC-min

− 2.6 mm [− 3.4 to − 1.8]

− 8.3 mm [− 9.7 to − 6.9]

3.0 mm [1.6 to 4.4]

0.79 [0.66 to 0.87]

M-SC

M-TH

IVCc

13.9% [9.5 to 18.3]

− 18.1% [− 25.7 to − 10.5]

45.8% [38.3 to 53.4]

0.08 [− 0.18 to 0.34]

IVC-max

− 1.7 mm [2.8 to − 0.6]

− 9.6 mm [− 11.5 to − 7.7]

6.1 mm [4.3 to 8.0]

0.53 [0.32 to 0.70]

IVC-min

− 4.4 mm [− 5.8 to − 3.0]

− 14.5 mm [− 16.9 to − 12.1]

5.6 mm [3.2 to 8.0]

0.38 [0.13 to 0.59]

AI-SC

AI-TH

IVCc

7.7%

− 19.2% [− 24.6 to − 15.1]

34.6% [30.4 to 40.0]

0.69 [0.52 to 0.81]

IVC-max

− 2.0 mm

− 9.2 mm [− 10.9 to − 7.9]

5.2 mm [3.9 to 6.9]

0.45 [0.21 to 0.65]

IVC-min

− 2.9 mm

− 10.0 mm [− 11.6 to − 8.8]

4.3 mm [3.1 to 5.9]

0.67 [0.49 to 0.80]

  1. In case of the IVC size analysis in M-mode (M), we analyzed a single measure which was the most reliable measure as decided by the experienced operator performing the calculations. In case of the analysis with artificial intelligence (AI), repeated measures were taken and saved in the database. Results of IVC collapsibility, minimum and maximum diameters (IVCc, IVC-min and IVC-max, respectively) are provided in terms of mean Bias and limits of agreement (LoA) with their relative 95% confidence interval (CI), where appropriate. We also provide intraclass correlation coefficient (ICC) to describe how strong the measurements resemble each other