The characteristics of the 15 patients are presented in Table 1. The mean age at baseline was 49.2 ± 12.7 years, with 14 (93%) patients being female. The median long-axis diameter of ASD was 25.5 mm (IQR, 16.5–31.3 mm). Eight patients had previously received PAH-specific therapy. WHO-FC I, II, and III symptoms were observed in 13%, 60%, and 27% of patients, respectively. None of the patients had WHO-FC IV symptoms.

Among the 15 patients, 7 (47%) achieved a PVR < 5 WU with PAH-specific therapy and were classified as the low-PVR group, whereas the remaining 8 (53%) were classified as the high-PVR group. The maximum baseline PVR in the low-PVR group was 9.5 WU. Partial anomalous pulmonary venous return was detected in a patient in the low-PVR group. Comparisons between the low-PVR and high-PVR groups revealed no significant differences in age (47.0 ± 12.2 vs. 51.1 ± 13.7 years, p = 0.55), ASD diameter (31.0 [IQR, 18.0–39.3] vs. 22.5 [IQR, 14.8–28.8] mm, p = 0.14), and BNP levels (99.6 [IQR, 52.8–566.1] vs. 97.9 [IQR, 65.7–123.7] pg/mL, p = 0.64). WHO-FC I, II, and III symptoms were observed in 14%, 43%, and 43% of patients, respectively, in the low-PVR group, and in 12.5%, 75%, and 12.5% of patients, respectively, in the high-PVR group, with no significant difference (p = 0.38).

Compared with the high-PVR group, the low-PVR group had significantly lower mean PAP (47.9 ± 3.9 vs. 58.5 ± 10.7 mmHg, p = 0.03), lower PVR (7.0 ± 1.3 vs. 13.9 ± 5.8 WU, p = 0.01), and higher pulmonary blood flow (Qp) (6.12 ± 1.14 vs. 4.12 ± 1.16 L/min, p < 0.01). Hemodynamic data at baseline and after PAH-specific therapy are presented in Supplementary Table 1. All patients received oral PAH-specific therapy. Within a median follow-up period of 5 months (IQR, 2.0–7.0 months), 14 patients (7 patients per group) were treated with endothelin receptor antagonists, all patients received phosphodiesterase type 5 inhibitors, and 13 patients (7 patients in the low-PVR group and 6 patients in the high-PVR group) received oral or inhaled prostanoids (Table 2).

In the total population, the mean PAP and PVR values significantly decreased from 53.5 ± 9.7 to 44.0 ± 11.8 mmHg (p < 0.001) and from 10.7 ± 5.5 to 7.0 ± 5.2 WU (p < 0.01), respectively, whereas Qp increased from 5.06 ± 1.52 to 7.27 ± 3.22 L/min (p < 0.01). Compared with the high-PVR group, the low-PVR group had a significantly lower mean PAP (34.3 ± 6.9 vs. 58.5 ± 7.7 mmHg, p < 0.01) and PVR (3.2 ± 1.2 vs. 10.3 ± 4.9 WU, p < 0.01) and significantly higher Qp (9.70 ± 2.72 vs. 5.15 ± 1.82 L/min, p < 0.01) after PAH-specific therapy. The mean PAP and PVR were significantly reduced after targeted PAH treatment in the low-PVR group, but not in the high-PVR group (Fig. 2A and B). The reduction rates in the mean PAP and PVR after targeted PAH treatment were significantly higher in the low-PVR group than in the high-PVR group (mean PAP: 28.7 ± 10.9% vs. 8.7 ± 13.6%, p < 0.001; mean PVR: 55.1 ± 16.1% vs. 22.5 ± 28.1%, p = 0.02).

Hemodynamic response to PAH treatment and AVT. A, B Changes in the mean PAP (A) and PVR (B) after PAH-specific therapy in the low-PVR and high-PVR groups. PAH-specific therapy significantly decreased the mean PAP and PVR in the low-PVR group. C PVR during NO challenge in AVT in the low-PVR and high-PVR groups. *p < 0.05, **p < 0.01, ***p < 0.001. PAP pulmonary arterial pressure, PVR pulmonary vascular resistance, PAH pulmonary arterial hypertension, AVT acute vasoreactivity test, PVR pulmonary vascular resistance, NO nitric oxide

AVT was performed in 13 patients (6 in the low-PVR group and 7 in the high-PVR group). There was no significant between-group difference in the change in mean PAP (12.3 ± 3.8% vs. 6.1 ± 9.2%, p = 0.15). However, the reduction rate in PVR during AVT was higher in the low-PVR group than in the high-PVR group (44.9 ± 8.9% vs. 15.4 ± 15.4%, p < 0.01, Fig. 2C). No adverse events occurred during AVT.

The baseline PVR (cut-off value = 9.5 WU, area under the curve = 0.93) and the change rate in PVR during AVT (cut-off value = 38.6%, area under the curve = 0.95) demonstrated good predictive accuracy (Fig. 3). PVR during AVT at cut-off values of 38.6% and 20% revealed sensitivities of 83.3% and 75%, respectively, with both showing 100% specificity. Compared with a PVR change rate of > 38.6% alone, the combination of a baseline PVR of < 9.5 WU and a PVR change rate of > 20% during AVT yielded higher predictive accuracy for response to PAH-specific therapy (100% sensitivity and specificity; Supplemental Fig. 1).

Predictors for response to PAH-specific therapy treatment. A Baseline PVR and PVR change rates during AVT in the low- and high-PVR groups. B Predictive power by cut-off value. A baseline PVR of < 9.5 WU and a PVR change rate of > 20% during AVT have 100% sensitivity, specificity, and accuracy. PAH pulmonary arterial hypertension, PVR pulmonary vascular resistance, AVT acute vasoreactivity test

In the low-PVR group, 6 of 7 patients underwent ASD closure (2 cases using device closure and 4 cases using surgical closure) without fenestration between 3 and 14 months after baseline (Table 3). Patients who underwent ASD closure did not experience hospitalization due to heart failure or death, and the improvement in hemodynamics was maintained. Furthermore, the burden of PAH-specific therapy decreased in this group; 2 of the 7 patients successfully discontinued all PAH medications, and 3 patients were able to reduce their dosages. The decision to taper or discontinue medications was made based on stable clinical status and hemodynamic data confirmed by follow-up examinations. Case 7, an older patient who did not undergo ASD closure due to a lack of consent, experienced hospitalization for heart failure followed by death.

Conversely, in the high-PVR group, all cases were treated with medication. All 8 patients required either maintenance or escalation of their PAH-specific therapies. During the follow-up period, PVR deteriorated in 4 of 8 patients (cases: 10, 11, 12, 13). Furthermore, 4 of the 8 patients (cases 8, 10, 11, and 15) required hospitalization due to heart failure, resulting in one mortality.