Logo image
Back
Non-Invasive Prediction of Pulmonary Hypertension Using Computerized Tomography
Abstract

Non-Invasive Prediction of Pulmonary Hypertension Using Computerized Tomography

Reshma Golamari, Asif Hameed, Harsh Bhatt, Sufana Shikdar, Shengnan Zheng, Abhishek Bharadwaj, Ajay Pillai, Oleg Teytelboym, Wei Yang, Aparna Babu, …
Circulation (New York, N.Y.), v 138(Suppl 1), 12824
06 Nov 2018

Abstract

Introduction Accurate diagnosis of pulmonary hypertension (PH) is made using invasive right heart catheterization (RHC). Prior studies have shown the utility of imaging techniques like computerized tomography (CT) as alternatives to predict PH.HypothesisTo determine the predictive ability of CT measured main pulmonary artery diameter (PA) to aortic diameter (AO) ratio in identifying PH. Methods A retrospective observational study conducted in 211 consecutive patients who had RHC and CT chest within 6 months of each other between the years 2012-2015. Two expert radiologists measured main PA diameter at the bifurcation and AO diameter from the same CT image. Furthermore, we conducted a sub-group analysis to see if the ratio performs better in assessing PH in chronic kidney disease patients. Linear correlation and logistic regression analysis was used to examine the association between PA:AO ratio by CT scan with invasive hemodynamics. Receiver operator characteristic analysis was used to evaluate the ability of using PA:AO ratio to predict PH. Resting PH was defined as mean pulmonary artery pressure of ≥25 mmHg. CKD was defined as GFR between 15- 60 mL/min/ 1.73 m and no indication for dialysis. Results Mean time between CT and RHC 3 days with minimum of 5.06 months (-152 days) to 5.56 months (167 days). Intraclass correlation coefficient between the two radiologists was 0.86 (95% CI of 0.75-0.92) with a P <0.01. CT measured PA:AO is highly correlated with mPAP after adjustment of age, race and gender (r=0.37, P<0.001). In a multivariate logistic model, every 0.15 U (1±SD) increase in PA:AO ratio was associated with 2.99 times higher (95% CI 1.99-4.49, P<0.01) odds of having PH , while in CKD subgroup, every 0.15 U (1±SD) increase in PA:AO ratio was associated with 11.9 times higher odds of having PH (95% CI 2.10-67.6, P =0.005), adjusting for age, race and gender. Conclusion CT derived PA:AO ratio is predictive of PH.

Metrics

11 Record Views

Details

Logo image