ARDS is a life-threatening pulmonary disease with a poor prognosis and an increased mortality rate[14, 15]. Such an adverse outcome may result from circulatory failure rather than hypoxemia[2]. Critical ultrasound examination is a rapid, non-invasive and reproducible operation, with a dynamic and visual presentation of the results, combining the monitoring results and diagnosis and treatment thoughts for critical disease. It plays an increasingly important role in the clinical diagnosis and treatment of ARDS.
Previous studies mainly focused on its diagnosis and treatment of ARDS, while the influencing factors for its prognosis and the value of ultrasonography for the prognostic evaluation have rarely been studied. It is of great significance to determine the actual application value of the auxiliary examination for the patients. First, there should be good reasoning for the patients requiring examination. Second, unnecessary examination and occupation of the medical resources should be avoided to ease the economic burden on the individuals, society and country. Therefore, continuous testing in clinical practice is needed for any new technique to determine its application value. The MIMIC-III database, established based on the electronic medical record, is a continuously updated dynamic data system, which reflects the diagnosis and treatment process of critically ill patients and has been commonly used by the researchers of intensive care medicine[16].
To the best of our knowledge, this is the first study assessing the effect of echocardiography on the prognosis of ARDS patients. In this work, patients undergoing TTE had a higher disease severity score and more comorbid conditions, suggesting a severer disease degree. It was found that the 28-d mortality rate of patients undergoing TTE significantly declined after adjusting the confounding factors. Several hypotheses were proposed to explain the survival benefit, and some variables were compared between the TTE and non-TTE groups. The infusion volume in the TTE group was smaller at day 1 and day 3 after ventilation. Vasopressors were used more often in the TTE group, probably because TTE promotes the timely use of vasopressors. Patients in the TTE group also stopped taking vasopressors earlier than those in non-TTE group, which could be related to several factors according to the analysis of PS matching. First, studies have shown that acute pulmonary heart disease occurs in 20-25% of ARDS patients[17], and many obstructive factors, including hypoxia-induced pulmonary vasospasm, hypercapnia, high airway pressure, inflammatory factor-induced vasoconstriction and lung volume collapse cause an increased pulmonary vascular resistance, which greatly affects the right heart function and pulmonary circulatory resistance[18]. During positive pressure ventilation, pulmonary vessels are compressed by the stretched alveoli, which leads to increased pulmonary artery resistance and obviously reduces the pulmonary circulation blood flow[19]. The increase in PEEP raises the pulmonary vascular resistance, which leads to right heart dysfunction and may eventually result in the occurrence and development of shock. Echocardiography assessment of the right heart function during treatment can reveal the major cause of hemodynamic involvement or instability, because the status of the right heart involvement greatly varies among different types of shock, which directly affects the development and implementation of the clinical therapeutic regimen. Second, in terms of fluid management, right heart enlargement can be caused by an acute increase in the blood volume. Acute right heart enlargement occurs when there is a failure of compensatory fluid discharge due to renal insufficiency or low MAP. This results in left ventricular diastolic restriction through the ventricular septum and the pericardium, increasing the left ventricular filling pressure, and thus the extravascular lung water[20].The fluid management strategy test of the ARDS Collaboration in 2006 well established that patients with ARDS can mostly benefit from the conservative fluid management strategy through shock correction (vasopressor dependence), keeping the circulatory stability and guaranteeing organ perfusion. Although the conservative strategy does not reduce the 60-d mortality, it can shorten the duration of mechanical ventilation and length of ICU stay and ameliorate oxygenation without increasing the incidence of other organ dysfunction[21]. Moreover, a systematic review and meta-analysis covering 2051 patients with sepsis and/or ARDS in 11 randomized trials in 2017 found no significant difference in the mortality rate between the restricted fluid management group and routine treatment group[22]. However, the ventilator-free duration was found to be significantly increased and the length of ICU stay was significantly shortened in the restricted fluid management group. Our study showed that the infusion volume was smaller in the TTE group, which may have contributed to the improved survival. Third, vasoactive drugs are an important treatment means to lower the pulmonary circulatory pressure. As the distribution of related receptors varies, attention should be paid to the different effects of vasoactive drugs on pulmonary circulation and systemic circulation. Generally, vasodilators may also affect the systemic circulation when they act on the pulmonary circulation, a contradiction that may be sharper in severely ill patients. On the one hand, vasodilators expand the pulmonary artery and lower the pulmonary circulatory resistance, which supports the restoration of the right heart function and reduces the central venous pressure (CVP). On the other hand, vasodilators reduce the systemic circulation pressure, leading to circulatory instability. In particular, when less obvious decline in PAP and obvious decline in the systemic circulation pressure are observed when vasodilators are used, the transseptal pressure will be altered, which results in a leftward shift of the interventricular septum, a significant decrease in the left ventricular end-diastolic volume and a decline in the cardiac output. As a result, the systemic circulation pressure is further decreased, leading to an autonomous vicious cycle of the right heart[23]. It has been confirmed that these drugs can be applied under the guidance of tricuspid annular plane systolic excursion(TAPSE), right/left ventricular area ratio and eccentricity index[17]. The above-mentioned discussion may include the reasons for the improved mortality of ARDS patients by ultrasonography.
In addition, the patients were divided into the Echo_2 group and non-Echo_2 group according to the time of ultrasonography. Unlike the Echo_1 group, patients in the Echo_2 group showed no improvement regarding the clinical outcome, with no significant difference in the 28-d mortality between the Echo_2 group and control group. It is believed that ultrasonography should be performed as soon as possible on ARDS patients to reduce the mortality rate to the largest extent. However, no relevant studies have been presented. Hence, larger-scale prospective randomized controlled trials should be performed in the future to determine the timing of ultrasonography for ARDS patients. Although early assessment will not necessarily help to avoid further lung injury throughout the course of the disease, it can enable us to adjust the ventilation strategy, thereby ameliorating the prognosis. Based on our experience, we can suggest that ALI or ARDS patients receive ultrasonography at the time of ICU admission, and regular review should be performed according to the disease condition. Echocardiography at admission can provide valuable information, not only about the current clinical conditions, but also about preexisting diseases (e.g., severe right ventricular hypertrophy suggests the presence of chronic lung disease). If the condition is stable, then echocardiography should be performed at least once a week before weaning and after extubation (i.e., fluid therapy monitoring). Moreover, echocardiographic can assist in determining the cause (or auxiliary factor) of the progressive respiratory failure, such as systolic pulmonary artery pressure(sPAP) elevation, right ventricular dilatation, progressive right or left ventricular failure or individualized treatment. Re-examination (ventilation and non-ventilation) should be performed whenever right ventricular dilation or dysfunction is developed (even by conventional echocardiography), other options (i.e., inhalation of nitric oxide and prone position) should also be considered, and patients should be monitored more closely using echocardiography.
Our study had some limitations. First, despite the large sample size of real data, the data came from a single medical center, and there may be deviations among the subjects in the medical level, habits and population. Second, this was a retrospective analysis, so a large amount of data might have been eliminated, and there might have been a selection bias due to the lack of key information and other reasons during data extraction. Although TTE is a non-invasive and convenient operation, it has a poor repeatability. We could not assess the consistency of TTE in this study, resulting in a measurement bias. Therefore, we suggest designing a prospective multi-center study based on similar studies to further observe the effects of TTE on patients with severe ARDS. Third, although the 28-d mortality rate was explored, some significant outcome variables were not considered in the analysis, including long-term mortality and ICU readmission. Finally, the MIMIC-III database included the cases before 2012. Some studies suggest expanding the Berlin criteria for ARDS to include patients who have undergone high-flow nasal oxygen therapy (at least 30 L/min) and meet other standards in the Berlin criteria[24], which may lead to a result deviation in this study.