The ability of echocardiography to predict long-term mortality after a submassive PE is not clear. Echocardiography is recommended during the acute management of PE to assess RV dysfunction so as to further stratify non-hemodynamically compromised patients into intermediate or low-risk prognostic categories . In the current study, we report for the first time that the day 1 RA/LA area ratio on echocardiography in patients with acute submassive PE predicted long-term mortality in both the derivation and validation cohorts. A day 1 RA/LA area ratio >1.0 on presentation conferred at least a three-fold increase in long-term mortality. This was independent of the expected predictive value of the Charlson Comorbidity Index and simplified PESI Scores.
Pulmonary embolism causes an increase in RV afterload. The RA and RV are thin walled and low pressure chambers, unlike the LV, and have limited capacity to compensate to the elevated RV wall tension. This results in RV dilation and dysfunction and worsening tricuspid regurgitation.
Previous studies have shown the presence of RV dysfunction post PE, in particular an increased RV/LV ratio, by both computed tomography pulmonary angiography (CTPA) and TTE, correlated with clinical outcomes and short-term mortality [32–37]. However in other studies, no clear association could be demonstrated [38, 39]. This discrepancy may be due to absence of an uniformly accepted standard for measuring the ventricular volumes on CT and the lack of electrocardiographic gating of the CTPA .
The RA/LA area ratio reflects the integration of multiple haemodynamic parameters including diastolic and systolic function of the RV, and filling volumes and pressures of the LA and LV. Amongst the general population without a history of pulmonary embolism, the RA/LA area ratio has been reported to be 0.81 ± 0.15 . A raised RA/LA area ratio has previously been shown by our group to correlate with the extent of pulmonary artery occlusion after PE . In acute PE, the sudden increase in RV afterload can lead to RV and RA dilation. This additionally impairs left- atrial and ventricular diastolic filling due to deviation of the inter-atrial and inter-ventricular septae leftwards. These changes have been observed on CT scans in patients with massive PE [42, 43]. A relative increase in RA and RV volumes coupled with decrease in LA and LV volumes leads to an increased RA/LA area ratio. Our results indicate that these changes are also evident in submassive PE without hemodynamic instability and are dynamic.
While it may be expected that RA/LA area ratio may relate to early outcome, the association with long-term outcome as demonstrated in the present study is novel. A particularly important finding of our study is that there is an acute recovery of the RA/LA area ratio and that this recovery does not predict a favourable long-term outcome. Therefore, TTE performed on or after day 2 of PE presentation in earlier studies are likely to underestimate the prognostic utility of the RA/LA area ratio. Pathophysiologically, our study indicates that it is the initial cardiac response to the thromboembolic insult and not the subsequent recovery that predicts long-term outcome. That the RA/LA area ratio is highly dynamic and responds acutely to volume and pressure change has been shown previously after percutaenous atrial septal defect closure. Within 24 hours of closure, Kelly et al. observed significant and immediate decrease in the RA/LA area ratio compared with baseline TTE, despite an extended period prior with significant right sided volume overload . This is due to the immediate volume and pressure unloading of the right heart, with complete normalization of the RA/LA area ratio at early follow-up. We postulate a similar mechanism occurs in patients with acute PE where right sided volume changes occur more acutely and therefore making these changes more reversible. Following the commencement of anticoagulation therapy and the resulting reduction in thromboembolic burden in the pulmonary vasculature, there is an immediate volume and pressure unloading of the right heart with improvement in the RA/LA area ratio within 24 hours. Presumably, these patients with elevated RA/LA area ratio on Day 1 indicate a reduced cardiac-pulmonary capacity to cope with increased RV afterload, even if there is normalisation of the RA/LA area ratio over the first week of treatment independent of the initial PE size as calculated by the simplified PESI score.
The RA/LA area ratio is simple to measure and a cutoff of day 1 RA/LA area ratio >1.0 may be a practical guide for physicians to identify patients who are at increased risk of death in the long-term following submassive PE. In particular, we have identified 17 of the 37 (46%) deaths in the validation cohort were cardiovascular or recurrent pulmonary embolus deaths which may have been preventable. These patients may benefit from long term follow-up and closer surveillance [6, 7] to ensure appropriate anticoagulation therapy is provided and that there is normalisation of the RA/LA area ratio and RV function with normalisation of pulmonary haemodynamics.
The CCI is a summation score of the burden of Comorbidities  shown to be very useful in prognosticating the outcome of patients suffering from diseases including heart failure , endocarditis  and cancer . In our current study, we confirmed the CCI is an independent predictor of long-term death following submassive PE. Among the included comorbidities in the CCI are cardiovascular diseases, chronic pulmonary diseases and malignancies. These comorbid illnesses have previously been found to be present in patients presenting with venous thromboembolism  and were predictors of long-term functional impairment and survival . We have shown for the first time that a simple and easily obtained echocardiographic parameter, the RA/LA area ratio, is an independent predictor of long-term survival following acute PE.
This study is limited by its sample size. This may explain why cardiac troponin-T and BNP levels, which were raised amongst the non-survivors, were not independent predictors of long-term mortality following multivariate analysis in our combined cohort. We have recently demonstrated a concentration-dependent relationship between cardiac troponin-T elevation following acute PE and long-term mortality  and it will be important to extend the present study into larger patient cohorts to directly compare RA/LA area ratio and biochemical parameters in predicting long-term outcome.