Early Critical Care Transthoracic Echocardiography Improves the Mortality of Patients Undergoing Mechanical Ventilation: Observational Data From Two Large Databases.

Background: Critical care transthoracic echocardiography (TTE) can quickly and accurately assess haemodynamic changes in ICU patients. However, it is not clear whether transthoracic echocardiography improves the prognosis of mechanically ventilated patients. In this study, we hypothesized that early critical care transthoracic echocardiography independently contributes to improvements in mortality in mechanically ventilated patients in the ICU. Methods: This was a retrospective study based on the Medical Information Mart for Intensive Care III (MIMIC-III) database and the eICU Collaborative Research Database (eICU-CRD). Patients undergoing mechanical ventilation for more than 48 hours were selected. The exposure of interest was early TTE. The primary outcome was in-hospital mortality. We used propensity score matching to analyse the association between early TTE and in-hospital mortality and sensitivity analysis, including the inverse probability weighting model and covariate balancing propensity score model, to ensure the robustness of our ndings. Results: A total of 8862 patients undergoing mechanical ventilation were enrolled. The adjusted OR showed a favourable effect between the early TTE group and in-hospital mortality [MIMIC: OR 0.77, 95% CI (0.63–0.94), (P=0.01); eICU-CRD: OR 0.78, 95% CI (0.68–0.89), (P<0.01) ]. Furthermore, TTE was also associated with 30-day mortality in the MIMIC database [OR 0.74, 95% CI (0.6-0.92), P=0.01]. Conclusions: Early application of critical care transthoracic echocardiography during mechanical ventilation is benecial for improving in-hospital mortality. Further investigation with prospectively collected data is required to validate this relationship.


Background
Mechanical ventilation (MV) is the cornerstone of supportive treatment for patients with acute respiratory failure and is a common and important procedure for patients in intensive care units (ICUs). There are 13 million to 20 million people undergoing mechanical ventilation per year worldwide [1]. However, the mortality of mechanically ventilated patients is still very high (28%-31%) [2,3]. As research continues, people have realized that the lung is not only a respiratory organ but also a haemodynamic organ [4].
During mechanical ventilation, positive pressure affects the patient's respiration as well as their haemodynamics. Although respiratory failure is the greatest problem to be solved in patients undergoing mechanical ventilation, haemodynamics also play a role that cannot be ignored [5], and a recent study also showed that haemodynamic changes caused by the in uence of airway pressure affect the prognosis of patients [6]. Thus, mechanically ventilated patients need to be monitored not only for respiratory function but also for haemodynamic stability [7,8].
Echocardiography, which can be conveniently performed in the ICU, provide more information on cardiac abnormalities, including anatomical abnormalities and functional abnormalities [9], and quickly and accurately assess haemodynamic changes in the ICU [10]. Understanding the clinical value of early transthoracic echocardiography (TTE) in mechanically ventilated patients is enormously important.
Several studies have suggested that cardiac assessment should be included in management strategies in patients undergoing mechanical ventilation. Transthoracic echocardiography is a safe, non-invasive method that can be used to assess the patient's uid response under mechanical ventilation [11] and is ideally suited to diagnose weaning failure of cardiac origin [12]. However, there is no strong evidence to support that transthoracic echocardiography should be early performed during mechanical ventilation [13]. The possible reason is that the research on transthoracic echocardiography mainly focuses on the management changes caused by transthoracic echocardiography, but the outcome impact of these changes is unclear [14]. Based on this question, our study was implemented to investigate the impact of earlier transthoracic echocardiography performance on the outcomes of critically ill adult patients undergoing mechanical ventilation. We hypothesized that thoracic echocardiography in the early stages of mechanical ventilation independently contributes to improvements in mortality in mechanically ventilated patients in the ICU.

Methods
This study was reported in accordance with the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) statement [15]. We collected data from Medical Information Mart for Intensive Care (MIMIC)-III v1.4 [16] and the eICU Collaborative Research Database (eICU-CRD) v2.0 [17]. Both are extensive, free, public databases containing hospitalization information. MIMIC covers 61,532 ICU admissions for 46,476 patients at the Beth Israel Deaconess Medical Center in Boston, MA, USA. The eICU-CRD covers 200,859 ICU admissions from 139,226 patients at 208 U.S. hospitals. We completed the required courses for the use of the database and obtained the corresponding certi cate (researcher certi cation number 1605699 and record id 27752407).

Study cohort
We conducted a retrospective study of mechanically ventilated adult patients from medical intensive care units (MICUs) and surgical intensive care units (SICUs) based on the method established by Serpa Neto et al [18]. We used only the rst ICU admission data for the rst hospitalization and patients 16 years of age or older who had been continuously ventilated for at least 48 hours. Patients who had incomplete datasets were excluded. Patients who underwent echocardiography less than 24 hours before mechanical ventilation or within 24 hours after mechanical ventilation were classi ed as the early TTE group, and the remaining patients constituted the group without earlier TTE (non-TTE).
Data were extracted from the database using structured query language (SQL). The following demographic data (using data from the rst 24 hours of admission) were collected: age, sex, weight, race, comorbidities (chronic obstructive pulmonary disorder [ we also collected management data for the rst day of mechanical ventilation (total IV uid; ventilator settings; use of dobutamine and norepinephrine).

Outcomes
The primary outcome of the study was in-hospital mortality. Secondary outcomes were 30-day mortality from the date of ICU admission; days free of mechanical ventilation and vasopressors 30 days after ICU admission; use of vasoactive drugs; total IV uid; and ventilator settings during the rst day of MV.

Statistical methods
To control for confounding factors, propensity score matching (PSM) was performed. The baseline characteristics of the original cohort were strati ed by TTE. The propensity score for an individual was determined based on the covariates age, sex, weight, race, HR, COPD, asthma, ARDS, sepsis, SOFA score, OASIS, WBC, Hb, pH, pO 2 , pCO 2 and lactate using a standard software package (matching package) with a PSM methodology. These variables were selected due to their clinical relevance. This method consisted of ranking the MV patients with TTE and non-TTE, then selecting the TTE patients who had the highest propensity score and nding the non-TTE patient with the closest propensity score (maximum calliper, 0.2). Both patients were then removed from consideration for matching, and the next highest patient was selected (matched 1:1 using the nearest-neighbor algorithm).
After matching, to assess the balance between the two groups, the standardized mean differences (SMDs) between the TTE cohort and the non-TTE cohort were calculated. SMDs eliminate not only the in uence of the absolute values from a study but also the in uence of the unit of measurement on the results [19]. Continuous variables are shown as the means and standard deviation, and categorical variables are represented as the total and proportion. For continuous variables, we used a nonparametric test or the Wilcoxon rank-sum test. For the categorical variables, we used a chi-square test or Fisher's exact test.
Secondary outcomes were observed after matching as well. We used paired t tests for continuous outcomes and chi-square tests for categorical outcomes.

Sensitivity analysis
We conducted a series of sensitivity analyses with the cohort with missing data, the cohort after imputation, and the cohort after PSM to assess the outcomes. In addition, we used multiple logistic regression, the inverse probability of treatment weight (IPTW) [21] and the covariate balancing propensity score (CBPS) [22] to further validate the primary outcome. To adjust for these covariates, the doubly robust estimation method [23] was used to deduce the independent associations between TTE and inhospital mortality and 30-day mortality (details about the IPTW and CBPS can be found in the Additional le 2). In addition, we used multiple logistic regression to analyse the impact of TTE during different time periods on the outcome (echo time I: patients who had TTE but the TTE time was not in echo time II; echo time II: TTE time >= MV time+24 hours and TTE<=MV time+24 hours). Finally, we carried out a sensitivity analysis through multivariate logistic regression focusing on patients with ARDS and sepsis.
Statistical signi cance was assessed to be determined by a two-sided p < 0.05. All statistical analyses mentioned above were performed using R version 3.5.3.

Results
After reviewing 46476 unique patients from the MIMIC-III database and excluding those with readmission, age<16 years, and ventilation duration<48 hours, and those receiving tracheostomy, 2790 patients from the MICU and SICU were enrolled (Fig.1). In the eICU-CRD, of the 139226 unique patients, 6076 patients in the MICU and SICU were included after the exclusion of patients aged<16 years, those receiving invasive ventilation for less than 48 h or tracheostomy and those missing hospital discharge information (Fig.1).
The in-hospital mortality of MV patients was 32.11% in the MIMIC database and 28.74% in the eICU-CRD. Patients who died were older and lighter and had higher OASIS and SOFA scores, higher lactate, and lower MAP (eTable.1).
The original cohort baseline in the MIMIC database showed that patients who underwent TTE on the rst day of mechanical ventilation had a more severe status in terms of SOFA score, MAP, pH, pO 2 , and pCO 2 (Table.1); however, there was no statistically signi cant difference between the TTE and non-TTE groups in the eICU-CRD (Table.1).

Secondary outcomes
After PSM, TTE was also associated with 30-day mortality in the MIMIC database [OR 0.74, 95% CI (0.6-0.92), P=0.01] (eFig.5). Since the eICU-CRD only contains in-hospital mortality data, we had no way to assess the association between TTE and 30-day mortality in the eICU-CRD cohort. In addition, we found that those who had TTE had both more ventilation-free days and more vasopressor-free days in 30 days than the non-TTE group (eTable.7), which might be related to the management changes brought by TTE, including the amount of IV uid (only in the eICU-CRD), the use of vasoactive drugs and the ventilator setting parameters (only in the MIMIC database). However, this conclusion should be regarded cautiously as far as the generalization of results is concerned.

Sensitivity studies
We performed some sensitivity analyses, as summarized in Fig.2 (in-hospital mortality) and additional le 1: eFig.5 (30-day mortality). We analysed all three cohorts, including the matched cohort, the original cohort with missing data, and the cohort after imputation, and found similar results: in-hospital mortality and 30-day mortality were improved in mechanically ventilated patients undergoing TTE. In addition, early TTE was more conducive to patient prognosis than TTE during other time periods (eFig.6), and the subgroup analysis also showed that TTE was bene cial in improving in-hospital mortality in the ARDS cohort (in the MIMIC sample only) and the sepsis cohort (eTable.5).

Discussion
In our study, the results provided evidence to support our hypotheses. After we adjusted for important confounding factors through PSM analysis, the results showed that TTE was associated with a reduced risk of in-hospital mortality and 30-day mortality in these patients. The above results were veri ed by the data from the MIMIC III and eICU-CRD databases. We also tested several hypotheses to account for the mortality bene t. It may be that TTE caused a change in management and improved the patient's prognosis, but based on current research, we cannot be completely sure.
We consider the following possible reasons for the improvement in in-hospital mortality and 30-day mortality with echocardiography: The lung is not only a ventilatory organ but also a blood-ow-related organ [4], and mechanical ventilation affects haemodynamics as well as the respiratory system [24]. The effect of ventilation on the haemodynamics of the heart is mainly due to changes in pleural pressure (Ppl), and Ppl affects both the in ow of the right ventricle and the out ow of the left ventricle [25]. While reducing the vena cava return, resulting in abnormal lling of the right ventricle, it will also increase the resistance of the pulmonary vascular vessels and affect the out ow function of the right ventricle. While affecting the right ventricle, it also affects the function of the left ventricle. These effects lead to acute cardiac strain and functional and even organic lesions. Intuitive assessment of cardiac function can help clinicians adjust ventilator settings to minimize the occurrence of cardiovascular dysfunction while maintaining ventilation [26].
TTE is a valuable tool for monitoring haemodynamics bedside [27]; it can easily provide dynamic haemodynamic parameters [28] and information about heart-lung interactions in mechanically ventilated patients [29]. It can comprehensively assess the structure and function of the heart and valves, stroke volume, ejection fraction, etc., provide physicians with speci c and direct information about left and right ventricular functions, and provide many other options to assess uid status or pharmacological demand [30,31]. It can provide the clinician with reliable information for making the correct clinical decision [11]. Recently, similar to our study, Dessap et al. found that early identi cation of diastolic dysfunction may allow the intensivist to design an intervention that might improve acute respiratory distress syndrome patient mortality [32], and Feng et al. also showed that early transthoracic echocardiography could improve the prognosis of sepsis patients by changing their management [33]. These ndings increase the possibility that TTE can provide physicians with useful information in the ICU. In addition, echocardiography is non-invasive and provides reliable information in most cases. Thus, although the effect of echocardiography on patient outcomes has not yet been established, experts recommend that patients undergo mechanical ventilation for echocardiography. They suggested that echocardiography should be performed early in the course of management to quickly obtain information on ventricular dimensions and function and to assess changes in cardiac output in response to therapy [4,28]. Our ndings provide evidence for the early use of echocardiography in mechanically ventilated patients and the possibility that TTE can be better applied in the ICU.
The present analysis had some limitations. First, since this was a retrospective study, to rule out the impact of professional physicians on the success of TTE, only patients from the MICU and SICU were selected. Second, although we tried to control the confounding factors as much as possible, the addition of other variables may have affected the results. Third, as this was a retrospective study, we cannot know whether the management differed between the TTE group and the non-TTE group with regards to TTE, and there is insu cient evidence to show that different management is associated with the mortality of MV patients. Fourth, we used PSM to adjust some of the potential selection biases. Although we performed a sensitivity analysis to make the results more reliable, some biases are inevitable. Finally, although our study suggested that the in-hospital mortality of MV patients was associated with TTE, further prospective randomized controlled trials are still needed to validate our results.

Conclusions
To the best of our knowledge, this is the rst report to determine the clinical value of TTE for patients undergoing MV. Early application of echocardiography during mechanical ventilation is bene cial for the prognosis of patients. However, further prospective, multicentre, randomized controlled studies are needed to validate our results. Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Data are reported as the mean (standard deviation) or no./total (%).
All data were extracted in the rst 24 hours of ICU admission. Figure 1 Flow chart illustrating the inclusion and exclusion criteria of the study.