The present results show that COVID 19-induced ARDS is associated with early and pronounced uncoupling of right ventricular function from the pulmonary circulation, and that its non-invasive echocardiographic assessment by the TAPSE/PASP ratio adds significantly and independently to the prognostic relevance of the PaO2/FIO2 ratio in these patients.
The reported COVID-19 patients were diagnosed with pneumonia complicated by ARDS based on clinical presentation of severe dyspnea, cough and fatigue, compatible chest computed tomography findings, and the PaO2/FIO2 ratio. At the moment of echocardiographic evaluation, the PaO2/FIO2 ratio had been corrected in a proportion of the survivors (Fig. 2). The patients were treated empirically with drugs expected to be of benefit, such as hydroxychloroquine and were anticoagulated. Their ventilatory management included proning, application of positive end-expiratory pressure and so-called "protective ventilation" with a low as possible tidal volumes (15). This resulted in a 26% mortality at the lower range of currently reported (17).
Pulmonary hypertension in the present study was mild to moderate as based on echocardiographic estimates of PASP. A PASP of 40 mmHg in the non-survivors would indeed be at the upper limit of normal taken into account age, sex, and body weight (18). On the other hand, the TAPSE was decreased but still above the lower limit of normal in the non-survivors (19). Accordingly, the TAPSE/PASP at 0.89 ± 0.29 in survivors was mildly decreased compared to the value of 1.11 ± 0.03 previously reported in 209 subjects older than 60 years (19). However, it was markedly decreased to 0.51 ± 0.22 mm/mmHg in non-survivors, approaching values below 0.50 mm/mmHg previously shown to be of poor prognosis in heart failure and severe pulmonary hypertension (12).
In a recent report of 200 hospitalized with COVID-19 in non-ICU departments, PASP was > 35 mmHg in 12% and the TAPSE < 17 mm in 14.5%, but increased PASP and not decreased TAPSE was found to predict a poor outcome (20). Mild pulmonary hypertensions along with moderate decrease in TAPSE in that study are in keeping with the present findings in more severely ill patients with respiratory insufficiency.
The TAPSE/PASP ratio was initially proposed as an estimate of RV myocardial length-tension relationship, and as such showed to be of prognostic relevance in heart failure (20). Subsequent studies confirmed its prognostic capability, not only in heart failure (21) but also in pulmonary arterial hypertension (22) and in patients with chronic lung diseases (23). In these studies, the TAPSE/PASP was assumed to inform about RV-PA coupling, with TAPSE considered as a load-dependent surrogate of Ees and PASP as an indirect estimate of Ea (21–23). The TAPSE/PASP has been shown to be superior to other composite echocardiographic indices in the assessment of RV-PA and correlated to gold standard invasive (11) or indirectly assessed Ees/Ea ratios. (21).
As in the present study the TAPSE/PASP ratio was mostly decreased in invasively ventilated patients, one could wonder if the application of positive end-expiratory pressure could have contributed to increased PAP and RV-PA uncoupling (24). COVID-19 ARDS patients could have presented with increased transmission of alveolar pressures to pulmonary resistive vessels because of preserved lung compliance (3). Mechanics of the respiratory system were not assessed in the present study. However, the notion of preserved compliance in COVID-19 ARDS may not be confirmed in most of these patients (25), and the "protective ventilation" approach in the present study would be expected to avoid to high volumes and alveolar pressures as a cause of excessive RV afterload (10). This was confirmed by only mild increases in PASP disclosed by the echocardiographic examinations.
The reason for RV-PA uncoupling in the presence of only mildly increased PAP is not immediately apparent. The basic response of RV function to increased afterload is homeometric, with increased Ees (contractility) to match Ea (afterload), and uncoupling expected but only in severe or rapidly evolving pulmonary hypertension (26). However, early RV-PA uncoupling may be observed in severe inflammatory conditions such as sepsis (27) or also in left heart failure because of negative ventricular interactions (28). Both may occur in COVID-19 patients (7). Therefore, the right heart in COVID-19 patients may fail even in the presence of only modest increase in afterload.
The present results are in keeping with a recent echocardiographic study in patients with COVID-19 ARDS, in which non-survivors had a PASP at the upper limit of normal, decreased indices of RV systolic function and longitudinal strain identified as an independent predictor of outcome (8). The TAPSE/PASP is easier to assess, can be part of standard bedside echocardiographic assessments as it does not require off-line analysis of images and specific software, and may be a more sensitive assessment of RV-PA coupling. The high prevalence of RV dilatation and dysfunction in the range of 40–50% recently reported in patients with COVID-19 (9, 29) underscore the exquisite sensitivity of the RV to this newly appeared viral infection.
The most potent predictor of outcome in ARDS is the PaO2/FIO2 ratio, which as such is part of the definition of the syndrome (13). In the present study, the TAPSE/PASP emerged with equally potent prognostic capability, suggesting a major component of acute cor pulmonale in COVID-19 ARDS pathophysiology. Whether this is entirely particular to COVID-19 ARDS is uncertain as there have been no systematic evaluations of RV-PA coupling in more "typical" ARDS or other viral pneumonia ARDS controls.
Approximately 4 decades ago, Zapol and Snider called attention to the pulmonary circulation and the right heart in severe ARDS (30). Pulmonary hypertension in these patients is nowadays uncommon along with progress in management, but "acute cor pulmonale" continues to be reported, albeit generally in the context of ventilatory settings associated with excessive increase in alveolar pressure and permissive hypercapnia (10). The present investigation shows that acute uncoupling of the right heart from the quasi-normotensive pulmonary circulation may also occur in the context of severe systemic inflammation and vasculitis.