Background
Methods
EIT
Study protocol
Off-line analysis
Statistics
Results
Age, years | 62.6 ± 14.8 |
Females/Males | 1/14 |
Smoker/Non-smoker, % | 38.5/61.5 |
SAPS II | 35 ± 10.6 |
Body weight, kg | 91.1 ± 10.3 |
Height, cm | 178.5 ± 7.2 |
BMI, kg/m2
| 28.8 ± 4.4 |
Heart rate, 1/min | 75 ± 16 |
MAP, mm Hg | 81 ± 11 |
Catecholamines, yes/no | 4/11 |
PaO2, mm Hg | 114 ± 42 |
PaCO2, mm Hg | 44 ± 8 |
SaO2, % | 97 ± 2 |
PF ratio, mm Hg | 255 ± 75 |
L1 | L2 | L3 | L4 | L5 | L6 | L7 |
P value | |
---|---|---|---|---|---|---|---|---|
RR, 1/min | 15.7 ± 3.6 | 15.5 ± 3.6 | 15.8 ± 3.8 | 15.7 ± 4.0 | 15.6 ± 3.7 | 15.9 ± 3.7 | 16.5 ± 3.9 | n.s. |
VT, ml | 649 ± 134 | 637 ± 141 | 627 ± 138 | 621 ± 132 | 611 ± 125 | 639 ± 154 | 607 ± 122 | n.s. |
VT PBW, ml/kg | 8.6 ± 1.8 | 8.3 ± 1.8 | 8.2 ± 1.9 | 8.1 ± 1.8 | 8.1 ± 1.8 | 8.3 ± 2.1 | 7.7 ± 1.5 | n.s. |
PEEP, mbar | 8.0 ± 2.4 | 8.1 ± 2.3 | 8.1 ± 2.3 | 8.1 ± 2.3 | 8.1 ± 2.3 | 8.1 ± 2.5 | 7.8 ± 2.3 | n.s. |
PIP, mbar | 20.7 ± 4.7 | 20.3 ± 4.7 | 20.2 ± 4.6 | 20.2 ± 4.6 | 20.2 ± 4.5 | 20.2 ± 4.8 | 18.5 ± 4.2 | n.s. |
Cres, ml/mbar | 83.3 ± 43.3 | 81.1 ± 35.7 | 74.7 ± 25.7 | 74.8 ± 26.7 | 75.9 ± 28.1 | 78.4 ± 30.4 | 83.2 ± 35.9 | n.s. |
Discussion
TV/VT ratio
Impact of Newton–Raphson algorithm and automatically adjusted and user-defined colour scales
EIT sensitivity region in previous studies
EIT sensitivity region at a juxta-diaphragmatic level
Clinical considerations
Spontaneous breathing
Gender
Conclusions
Key messages
-
The clinical usability and plausibility of EIT measurements depend on proper belt positioning, proper impedance visualisation, correct analysis and data interpretation.
-
Clinical users should apply the electrode belt at ICS 4–5 (parasternal line).
-
If an out-of-phase signal or and overshoot signal exist, its origin should be clarified. If this signal is caused by a juxta-diaphragmatic belt position, it should be eliminated by repositioning the belt.
-
Data containing out-of-phase signals should not be used for the analysis of ventilation-related impedance changes.
-
If existing analysis tools are applied (e.g., “auto-scaling”) and if users do not take resultant effects into account, their findings might be biased and drawn conclusions are potentially wrong.