Prediction of Balloon Pulmonary Angioplasty Response for Chronic Thromboembolic Pulmonary Hypertension Using Fractal Analysis of 99mTc-MAA SPECT

Objectives Chronic thromboembolic pulmonary hypertension (CTEPH) is a life-threatening disease, which results in pulmonary hypertension due to stenosis or obstructions of the pulmonary arteries and the prognosis of CTEPH patients with mean pulmonary arterial pressure (mPAP) more than 30 mmHg is poor. Balloon pulmonary angioplasty (BPA) has been recently reported to improve hemodynamic parameters and symptom in inoperable CTEPH, however, noninvasive quantitative tool for therapeutic effect of BPA is not still established. The purpose of this study was to propose noninvasive quantitative index in BPA response for CTEPH using three-dimensional (3D) fractal analysis of 99mTc-MAA SPECT.

Methods Data of pulmonary perfusion 99mTc-MAA SPECT for 29 patients with CTEPH who underwent BPA between 2013 and 2015 was analyzed. According to 3D fractal analysis, the fractal dimensions were calculated from the relationship between the 5 cutoff levels (30%, 35%, 40%, 45% and 50% of the maximal voxel radioactivity) and the perfusion volume (ml) surrounded by the 5 cutoff levels on the whole lung of SPECT images transformed into natural logarithms. The fractal dimension was used as heterogeneity in pulmonary perfusion. In addition, the right and left perfusion volumes (RPV and LPV) were measured as the total areas surrounded by the contours of the cutoff level of 30%. After BPA, the CTEPH patients were divided into patients with mean pulmonary arterial pressure (mPAP) < 30 mmHg (higher mPAP group) and patients with mPAP 蠅 30 mmHg (lower mPAP group). Comparison of RPV, LPV, and the fractal dimension between pre- and post-BPA was analyzed by the paired t-test. Comparison of the parameters between higher and lower mPAP group was analyzed by the Mann-Whitney U-test. The predictability of higher mPAP group was performed by receiver-operating-characteristic (ROC) analysis.

Results RPV was significantly greater at post-BPA than pre-BPA (366 ± 142 ml vs. 218 ± 182 ml, p <0.05). There was no significant difference in the LPV and the fractal dimension between pre- and post-BPA. The fractal dimension at post-BPA was significantly greater for higher mPAP group (n=15, 2.7±0.4) than lower mPAP group (n=14, 2.2±0.4, p <0.05). The increase of RPV and LPV after BPA was significantly greater for lower mPAP group than higher mPAP group (RPV; 221 ± 176 ml vs. 84 ± 222 ml, LPV; 37 ± 165 ml vs. -72 ± 82 ml, p <0.05). Use of optimal cutoff thresholds of fractal dimension at post-BPA 2.8 and increase of LPV 22 ml differentiated higher mPAP group from lower mPAP group, with 53% and 93% sensitivities, 100% and 71% specificity, and areas under the curve of 0.77 and 0.77.

Conclusions CTEPH patients with lower mPAP showed greater increase of perfusion volume and less heterogeneity in the whole perfusion than those with higher mPAP. 3D fractal analysis is a feasible and noninvasive technique to evaluate BPA response for CTEPH.