In this study of diagnostic accuracy, the investigators aim to validate a faster, simpler, and noninvasive test of fluid-responsiveness in critically ill children. This test is based on an assessment of the hemodynamic effects of a standardized abdominal compression, using electrical cardiometry. This would help physicians to identify more easily which patient could benefit from a volume expansion, thus avoiding a potentially useless or even dangerous fluid expansion that could lead to fluid overload. To this end, the diagnostic accuracy of electrical cardiometry-based stroke volume (SV) variation induced by a standardized abdominal compression to predict fluid responsiveness (define as a 15% increase in echocardiographically measured SV after volume expansion) will be measured.

Volume expansion is the cornerstone of acute circulatory failure treatment in children. However, inappropriate administration can lead to fluid overload, which is associated with poor outcome. Thus, the search for indicators to predict fluid responsiveness is a major issue in pediatric intensive care unit. In such an emergency context, this assessment must be as simple and fast as possible, and ideally non-invasive. In children, respiratory variability of peak aortic velocity is the most studied test and has an excellent diagnostic accuracy but is only validated in the absence of any spontaneous respiratory movement, a rare situation in practice. Recently, the abdominal compression maneuver has been investigated. This classical clinical maneuver induces a transient and reversible preload increase by increasing the venous return via the mobilization of the hepato-splanchnic venous reservoir. The echocardiographic evaluation of this maneuver can accurately predict fluid responsiveness. However, this echocardiographic assessment is a pitfall, as it is an operator-dependent, discontinuous and time-consuming examination. Therefore, electrical cardiometry could be interesting. This continuous and non-invasive cardiac output monitor could be an interesting alternative to evaluate the hemodynamic effects of the abdominal compression maneuver in order to predict fluid responsiveness.

Therefore, the investigators will evaluate the diagnostic accuracy of an abdominal compression maneuver for the diagnosis of fluid responsiveness in critically ill children, for whom the physician in charge prescribed a 10ml/kg volume expansion for circulatory failure. The index test will be the SV variation induced by an abdominal compression maneuver, with SV being measured with electrical cardiometry. Fluid responsiveness will be defined as a > 15% increase in echocardiographically-measured SV between baseline and within one hour after fluid expansion (gold standard test).

In this non-interventional, prospective, multi-center study of diagnostic accuracy, children requiring a 10ml/kg volume expansion will be included. After a prescription of 10 ml/kg volume expansion by the physician in charge, screening for inclusion and exclusion criteria will be performed by an investigator physician. If the patient is included, volume expansion will be delayed for a few minutes (<3 min) while an echocardiography and the index test are performed, but no supplemental blood test or invasive parameters will be collected. Another echocardiographic evaluation will be conducted within one hour after volume expansion, to assess response to fluid expansion (gold-standard test). Patients will be follow-up until PICU discharge (28 days maximum).

The index test is the ΔSV-ACICON: percentage of SV variation between baseline and during a standardized abdominal compression. Stroke volume will be assessed by electrical cardiometry, a noninvasive cardiac output monitoring.

The abdominal compression maneuver, a common clinical test in critically ill children, will be standardized as follow: a sphygmomanometer will be inflated with 80ml of air and will be applied to the center of the patient's abdomen. Then, the operator's hand will be placed on the sphygmomanometer (which is thus interposed between the patient's abdomen and the operator's hand) and a gentle manual anteroposterior compression will be performed, calibrated at 30 cmH2O according to the sphygmomanometer. Parameters of interest will be collected 30 seconds after the start of the compression.

Patients included in this study would have received volume expansion anyway, as the prescription of 10ml/kg volume expansion by the physician in charge is the main inclusion criterion. The purpose of this study is to evaluate the diagnostic accuracy of a fluid responsiveness test. To this end, patients:

• Will undergo echocardiographic assessments, which is noninvasive, non-radiative, well tolerated and commonly used in this population as a standard of care in our center.
• Will undergo a standardized abdominal compression, a commonly used maneuver in our center and in other pediatric intensive care units, which differs only slightly from a standard physical examination.
• No additional blood test or invasive parameters will be collected be collected. Number of subjects: Forty-two patients are needed. This was calculated with an alpha risk of 0.05, an expected prevalence of 0.5, a predicted area under the receiver operating characteristics curve of 0.90 (based on previous studies) and a width of the 95% confidence interval narrow enough to exclude 0.80 which is a threshold for clinical relevance.

Statistical analysis: Comparisons between fluid "responders" and "non-responders" will be performed with Student's t-test or Mann-Whitney test for continuous variables; or with the Chi-2 test or Fisher's exact test for categorical variables. Spearman correlation coefficient will be used to test linear correlations.

Fluid responsiveness represents the gold standard test. Patients will be classified as fluid responders in case of at least 15% increase in SV after fluid expansion, compared to baseline.

Diagnostic accuracy of the index tests will be explored. As multiple cut-offs of theses index tests can be defined, the investigators will report a receiver operating characteristic (ROC) curve which graphically represents the combination of sensitivity and specificity for each possible test positivity threshold. The area under the ROC curve and its 95% confidence interval informs in a single numerical value about the overall diagnostic accuracy of the index test. The ROC curves of all index tests will be compared using a DeLong test.

A p value less than 0.05 will be considered statistically significant

Inclusion Criteria:

• Age less than or equal to 8 years old
• Hospitalization in a pediatric intensive care unit
• Prescription by the attending physician of a fluid expansion of 10ml/kg
• Use of a non-invasive cardiac output monitoring device (electrical cardiometry) as part of routine care

Exclusion Criteria:

• Patient less than 37 weeks' corrected gestational age
• Hemodynamic instability making the delay necessary for any test dangerous
• Supine position contraindicated or deleterious
• Impairment of echocardiographic acoustic window or restless patient making ultrasonography impossible
• Opposition to participate expressed by the patient or by a parent or legal guardian
• Intra-abdominal hypertension, painful abdominal palpation or abdominal surgery in the last 15 days

patient hospitalised in intensive care or in a continuous monitoring unit