Sepsis shock is a leading cause of death in children globally. The mortality rate of septic shock is as high as 25%, despite advanced supportive care (1). Currently, fluids therapy remains the cornerstone of the hemodynamic resuscitation of children with septic shock. Fluid resuscitation seeks to rapidly restore the effective circulating blood volume and oxygen delivery to organs. Effective fluid resuscitation can improve the prognosis (2, 3), while repeated or inappropriate fluid bolus administration is associated with increased mortality and the length of intensive care unit (ICU) stay (4). Previous studies have shown that only 40–69% of children respond to intravascular volume expansion (5–8)_. Fluid overload is a common complication during the resuscitation of septic shock and can be avoided by predicting fluid responsiveness. Therefore, we have challenges in identifying which patients will respond to volume expansion.

Numerous hemodynamic variables could be predictors of fluid responsiveness (9, 10). Traditional static pressure indexes such as central venous pressure (CVP) or mean arterial pressure (MAP) had a limited predictive value for fluid resuscitation (11, 12). Dynamic indices have shown to be good predictors of fluid responsiveness in recent years, especially the indices relying on heart-lung interactions, such as inferior vena cava respiratory variability index, velocity-time integral (VTI) in left ventricular inflow tract, stroke volume variation (SVV), and so on (13–15). VTI or SVV was able to reliably predict the fluid loading response and was superior to traditional pressure indexes for adults (16). Children may have different patterns of fluid responsiveness under critical care conditions. The predictive value of these dynamic variables remains unclear (5, 17). Apart from this, VTI and SVV monitoring need highly skilled operators to ensure measurement accuracy and require specialized pieces of equipment that are not available in every hospital. These make them more difficult to perform in the clinical environment. Thus, more ICU physicians are willing to choose the inferior vena cava (IVC) as a point-of-care parameter of volume status assessment due to their convenience for measuring. A systematic review and meta-analysis found respiratory variations in the inferior vena cava diameter performed moderately well in predicting fluid responsiveness in adult patients with circulatory shock receiving mechanical ventilation (18). However, the value of IVC in predicting fluid responsiveness among pediatric populations was not clear.

Thus, we performed the single-center prospective observational study to evaluate the predictive utility of respiratory changes in IVC diameter for fluid responsiveness in children with septic shock undergoing mechanical ventilation as well as to provide evidence-based clues for clinically reasonable fluid therapy in these patients.

This study was undertaken to identify the respiratory variations in IVC diameter as a predictor of fluid responsiveness. Our results show that ΔIVC at a cutoff value of 28.5% can be used to predict fluid responsiveness in mechanically ventilated children with septic shock. It appeared that the ΔIVC was more effective in the prediction of the fluid responsiveness when compared to the other indicators, such as dIVC, CVP, MAP, and HR.

Currently, two major approaches are being used for assessing fluid responsiveness: the static and dynamic approaches. In this study, we found a significant increase in CVP in both R and NR groups after volume expansion (p < 0.01). Elevated CVP may be related to intravenous fluids treatment before admission. The area under the ROC of CVP was only 0.549. The result showed that CVP had no significant diagnostic value for predicting fluid responsiveness. This is consistent with previous reports that CVP, which is a classical static hemodynamic variable, cannot reliably predict fluid responsiveness (12, 28). A meta-analysis study (29) also confirmed that CVP can be regarded as an indicator of right ventricular end-diastolic volume index and can therefore not be regarded as an indicator of preload responsiveness to guide fluid therapy. Meanwhile, MAP did not indicate a significant predictive value in evaluating fluid responsiveness in our study. After volume expansion, the different amplitude of elevations in MAP was seen in both groups. Blood pressure is mainly determined by stroke volume and arterial elastance. The magnitude of the BP elevation caused by an increase in stroke volume may be partially offset due to a lower arterial elastance in children. Although these changes in MAP in fluid responders were statistically, MAP is still not a reliable indicator of predicting volume.

The most important treatment strategy for patients with septic shock is initial fluid resuscitation to restore hemodynamic stability. According to the Frank-Starling principle, increasing the heart preload can significantly improve the cardiac output only when both the left and right ventricles are in the elevated phase of cardiac function. It has been unclear as to which patients are volume-responsive and likely to benefit from fluid resuscitation. Although fluid resuscitation therapy is crucial in the management of hemodynamically unstable critically ill patients, excessive or inadequate fluid resuscitation can be harmful. A similar finding was shown in our study. Of the 58 patients with septic shock, 31 patients (53.4%) were non-responsive to volume expansion. In the NR group, it did not show a significant increase in cardiac output or a significant improvement in the heart rate and blood pressure after volume expansion. Judging the patients' volume status remains challenging.

Bedside ultrasound is an important method of evaluating the volume status of critically ill patients and has been widely adopted in intensive care units. Moreover, point-of-care ultrasound is a convenient and non-invasive method to perform hemodynamics measurements and hemodynamic monitoring (30). This hemodynamic information would be useful for the implementation of goal-directed hemodynamic therapy and help to tailor better fluid therapy protocol. Compared with other echocardiographic measurements, such as SV, VTI, and aortic peak blood flow velocity variation rate of breathing, the measurement of IVC diameter is relatively easy. The cardiac output may induce cyclic changes in intrathoracic pressure, especially in patients undergoing mechanical ventilation under deep sedation. Hence, there is a significant increase in the percentage of IVC diameter change in patients with volume deficits. Some studies (17, 31) suggest that respiratory change in IVC diameter is a useful predictor of fluid responsiveness. However, a systematic review and meta-analysis (32) stressed that ultrasound evaluation of the diameter of the IVC and its respiratory variations in fluid responsiveness had extreme heterogeneity. These results were inconsistent. Several factors affect the accuracy of the IVC for volume status judgments: the measurement sites of inferior vena cava diameter (33), the skill of the operators, the heart function, abnormal intrathoracic pressure, intra-abdominal hypertension, and so on.

Therefore, more stringent inclusion and exclusion criteria were developed in our study. We enrolled children with septic shock who received controlled mechanical ventilation under sedation and analgesia to extensively eliminate the effects of voluntary breathing and improve mechanical synchronism. Moreover, we included patients with moderate PEEP levels and a tidal volume ≤ 10 ml/kg to minimize the effects of PPEP and intrathoracic pressure on the diameters of IVC. Eventually, we found that the cutoff value of ΔIVC (before volume expansion) for predicting fluid responsiveness was 28.5%. Kutty et al. (34) reported that IVC collapsibility varied with body surface area (BSA), and the average was above 30% in normal pediatric subjects since this average value is higher than the 28.5% cutoff in this study. The interpretation of the inconsistent results can consider the following factors: intrathoracic pressure changes were highly regulated and the influence of spontaneous breathing is maximally removed. Furthermore, we found that the sensitivity of a ΔIVC > 28.5% to predict fluid responsiveness was 0.954 with a specificity of 0.685. While the sensitivity of ΔIVC was good, the specificity was not high. A voluntary inspiratory effort may explain in part the relatively low specificity of ΔIVC. The respiratory variation of the IVC diameter will increase accompanied by strong spontaneous breathing. However, we were not able to confirm whether fully eliminating the work of breathing and the spontaneous breathing efforts had no influence on the respiratory variation of the IVC diameter in our study. In addition„ other factors could also affect the relatively low specificity of ΔIVC: intrathoracic pressure changes, CVP, lung compliance, airway resistance, and so on (35, 36).

Moreover, in this study, we found that dIVC was not the significant predictor associated with volume responsiveness. Saritaş et al.'s (37) study showed the dIVC had a more accurate predictive role in predicting the volume status when compared with the ΔIVC among the patients with spontaneous respiration receiving different positive pressure support. Another research (38) showed the IVC area distensibility index and its diameter ratio in cross-section had more value than the IVC diameter distensibility index for predicting fluid responsiveness in mechanically ventilated patients. Thus, it still needs further research to confirm the value of dIVC.

This study found that the ΔIVC was a useful predictor with the ability to predict responsiveness in mechanically ventilated children with septic shock. The result was obtained in a selected patient population treated by specific protocols. The not so high specificity of ΔIVC suggests that clinicians should have a comprehensive evaluation of the ΔIVC in conjunction with the patient's underlying disease status and other hemodynamic changes, thereby, contributing to clinical fluid management.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Chengdu Women's and Children's Central Hospital in Sichuan, China [Registration No. 2017(11)]. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

Study design and Project administration: GZ and YQ. Acquisition of data: ZX, QZ, GZ, and BL. Formal analysis: ZX and MW. Writing original draft: ZX and XZ. Writing review and editing: ZX, YQ, and MW. Funding acquisition: YQ. All authors have read and agreed to the published version of the manuscript.

This work was supported by the National Natural Science Foundation of China (81771634 and 81971428) and the Fundamental Research Funds for the Central University (SCU2020D006).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.