Researchers Characterize Limitations With Methods of Determining IVS Flattening in PH

Researchers of a new study are underscoring the value of automated measures for determining the presence of interventricular septal (IVS) flattening in patients with pulmonary hypertension (PH).1

Findings from the retrospective study suggest that both visual and quantitative measures have limitations in identifying IVS flattening. IVS flattening is a critical part of echocardiographic measurement. IVS flattening describes the dividing wall between left and right ventricles appearing flat on an echocardiogram, which signals the increased amount of pressure or volume overload in the right ventricle and indicates PH. The presence of IVS flattening has been associated with worse outcomes in several cardiovascular conditions, including severe tricuspid regurgitation and pulmonary arterial hypertension.2,3

“Despite its incorporation into PH guidelines and clinical care, little is known about the reliability, accuracy, and test characteristics of the visual assessment and quantification of IVS flattening,” described the researchers. “To our knowledge, only 1 prior study described inter-observer agreement between echo readers for IVS flattening in 60 patients with PAH, demonstrating a good agreement with a Kappa of 0.78.”

This prior study, explained the researchers, was limited by incomplete details on the patient characteristics, the echocardiogram readers, and the description of how IVS flattening was determined. The study was also limited by a smaller cohort size, with 60 patients included in the analysis.

Automated measures can be used to find the presence of interventricular septal flattening in patients with pulmonary hypertension | Image credit: appledesign - stock.adobe.com

The new study, published in Respiratory Medicine, included 173 patients from a single center in the United States. The presence of IVS flattening was determined visually by 2 independent board-certified echocardiographers as well as by eccentricity index (EI). An EI cutoff of 1.5 showed the best combination of specificity (80%) and sensitivity (32%) to distinguish from visual IVS flattening.

Compared with the quantitative measure, visually determining IVS flattening showed high specificity but significantly lower sensitivity. When the researchers used an EI cutoff of at least 1.1 at end-systole to determine the metric, visual assessment had a sensitivity of 39% (95% CI, 23%-36%) and specificity of 85% (95% CI, 77%-90%). In the context of EI, visual assessment also had a negative predictive value of 44% (95% CI, 46%-57%) and positive predictive value of 73% (95% CI, 63%-80%), suggesting that the approach may lead to IVS flattening going unrecognized in many patients.

However, the researchers also found that EI failed to show a strong correlation between the 2 readers, with a 95% limits of agreement (–0.72 to 1.23) based on Bland-Altman analysis.

Between the 2 echo readers, the researchers found overall agreement but differing levels of agreement within subgroups of patients. Overall, the percent agreement for IVS flattening of 72%, with a κ coefficient of 0.27 (P = .0001). Agreement was considered moderate for the 57 patients with precapillary PH (80% agreement; κ = 0.51; P = .0001), was considered fair for the 55 patients with combined pre- and postcapillary PH (Cpc-PH) (73% agreement; κ = 0.31; P = .007), and was considered poor for the 61 patients with isolated postcapillary PH (66% agreement; κ = –0.07; P = .83).

“Interestingly, the level of agreement in visual IVS flattening between echo readers seemed to differ based on pulmonary vascular resistance, with highest agreement for patients with pre-capillary and Cpc-PH, which are both characterized by an elevated [pulmonary vascular resistance],” detailed the researchers. “In contrast, there was a lack of agreement in visual IVS flattening for patients with isolated post-capillary PH. This may have relevance when incorporating IVS flattening in echo scoring systems used to distinguish between PAH and LHD-PH.”

References

1. Rahman M, Jakkoju A, Mohsen AM, et al. Inter observer reliability and accuracy of the visual assessment of interventricular septal flattening in pulmonary hypertension. Respir Med. Published online November 13, 2024. doi:10.1016/j.rmed.2024.107858
2. Omori T, Maeda M, Kagawa S, et al. Impact of diastolic interventricular septal flattening on clinical outcome in patients with severe tricuspid regurgitation. JAHA. 2021;10(20):e021363. doi:10.1161/JAHA.121.021363
3. Ghio S, Klersy C, Magrini G, et al. Prognostic relevance of the echocardiographic assessment of right ventricular function in patients with idiopathic pulmonary arterial hypertension. Int J Cardiol. 2008;140(3):272-278. doi:10.1016/j.ijcard.2008.11.051