Right Heart Dysfunction in Adults with Coarctation of Aorta: Prevalence and Prognostic Implications

Egbe AC, Miranda WR, Jain CC, Connolly HM. Circ Cardiovasc Imaging. 2021 Dec;14(12):1100-1108. doi: 10.1161/CIRCIMAGING.121.013075. Epub 2021 Dec 8. PMID: 34875855

 

Take Home Points:

  • Right heart anatomic and hemodynamic abnormalities were evaluated by echocardiography in 821 patients with native (176) or repaired (645) coarctation who presented for evaluation between January 2000 and December 31, 2018. Median follow up was 8.2 years.
  • Right heart anatomic and hemodynamic abnormalities were present in almost 20% of the population; all patients studied had coarctation gradients < 20 mmHg but over 50% had hypertension.
  • Cardiovascular events (hospitalization for heart failure = 9%; transplant = 1%; cardiovascular death = 10%) occurred in 54 (14%) patients.
  • Using multivariate analysis, the authors determined that abnormalities in 4 indices – RA strain, RA volume index, RV global longitudinal strain and RV systolic pressure – were highly associated with cardiovascular events. The authors used these indices to generate a Right Heart Hemodynamic Score (RHHS), or risk scores, to help prognosticate in these patients.

Commentary from Dr. Thomas Zellers (Dallas, USA), section editor of Congenital Heart Disease Interventions and ACHD Journal Watch:

Patients with left heart obstruction develop LV systolic and diastolic dysfunction which can lead to pulmonary remodeling, pulmonary hypertension and RV dysfunction. Echo is capable of evaluating right heart anatomic and hemodynamic dysfunction and these indices are used for prognostication in patients with left heart disease. Coarctation of the aorta is known to lead to left heart disease and early coronary disease but there is limited data on its effect on right heart indices and their prognostic significance in this population.

 

These authors studied 821 patients who presented between January 2000 and December 2018 with the diagnosis of coarctation, as an isolated disease (n=563) or in combination with LV outflow (n= 204; aortic stenosis, subaortic stenosis, supra-aortic stenosis with mean gradient > 20 mmHg or > 2+ AI) or LV inflow (n=54; mitral stenosis > 3 mmHg or regurgitation > moderate) disease. Mean age was 32 (21-46) with 58% males. Half of the patients had hypertension; beta blockers were prescribed 29% of the time, calcium channel blockers 13% of the time and RAAS antagonists and diuretics were prescribed 28% and 11% respectively. They further validated their studies by dividing the group into 2 random cohorts, the derivation and validation cohorts (n= 411 and 410) and compared the right heart indices and cardiovascular outcomes. Median follow up was 8.2 years in the derivation cohort.

 

The authors used echocardiography with speckle tracking strain imaging to evaluate the right heart including RA volume index, RA reservoir strain, RA pressure estimates, RV end diastolic area, RV global longitudinal strain, tricuspid regurgitation and estimates of RV and PA systolic pressures. These seven indices were used to generate a right heart hemodynamic score (RHHS) based on univariate and multivariate analysis.

 

The primary outcome was a composite of cardiovascular events, including hospitalization for heart failure, transplant or cardiovascular death.

 

Right heart indices were abnormal in approximately 20% of the patients as follows: RA dysfunction 16%, RA volume enlargement 28%, RA hypertension 17%, RV systolic dysfunction 14%, RV enlargement 9%, at least moderate tricuspid regurgitation in 5% and pulmonary hypertension in 20%. All indices were worse in patients with LV inflow and LV outflow disease compared to isolated coarctation. There was also a statistical correlation between RV systolic pressure/pulmonary hypertension and the findings of RV longitudinal strain and RA reservoir strain.

 

The derivation and validation cohorts were used to compare RHHS and subsequent model fit. In the derivation cohort of 411 patients, 9% were hospitalized for heart failure, 1% required transplant and 10% died from cardiovascular disease or its associated complications for a total of 54 patient (14%) with cardiovascular events. All seven right heart indices were associated with endpoint cardiovascular events on univariate analysis; four indices (RA strain, RA volume index, RV longitudinal strain and RV systolic pressure) remained independently associated with cardiovascular events on multivariate analysis and the RHHS was generated (0-5) using these 4 variables. RA reservoir strain, RA volume index and RV global longitudinal strain were each assigned 1 point, and RVSP > 40 mmHg was assigned 2 points. The patients were stratified based on their RHHS into low risk (RHHS 0-1), moderate risk (RHHS 2-3) and high risk (RHHS 4-5). Moderate and high risk scores were more likely to be associated with inflow and outflow disease and older age, but not with age at repair. Residual coarctation mean gradient did not correlate with right heart indices. In addition to the right heart indices, LV global longitudinal strain, atrial fibrillation and hypertension (even with low coarctation gradients) were independently associated with cardiovascular events. When the two cohorts, the derivation and validation cohorts, were compared, the findings were similar between the two groups and the C-statistics for the models suggested a tight fit (C-statistics of 0.72 and 0.71, respectively).

 

Overall, right heart indices were abnormal in almost 20% of patients in the study group. Abnormal right heart indices and RHHS were associated with cardiovascular events and can be used for prognostication in this group. Interestingly, residual coarctation or time from repair at the time of evaluations had no prognostic significance suggesting that the abnormalities seen in the left heart leading to right heart abnormalities may be longer standing. Looking earlier for these abnormalities and/or earlier proactive/preventive therapy studies are the next steps to evaluating cause and prevention.

 

Limitations: retrospective study, no invasive hemodynamic indices and no evaluation of medical interventions to alter outcomes.