Evidence for Pulmonary Vascular Disease Despite Absence of Overt Pulmonary Arterial Hypertension (PAH) in Myeloproliferative Disorders (MPD) (PV and ET).

Background: Until recently, the association of PAH unrelated to thromboembolic events with MPD was occasionally reported, essentially in patients with myelofibrosis. In a large retrospective study, 26 cases of unexplained PAH associated with MPD (including 6 PV and 5 ET) were identified in the referral population of the Mayo Clinic (

Methods: MPD diagnosis (WHO criteria) was polycythemia vera (PV) (n=25), and essential thrombocythemia (ET) (n=20). 9/11 patients tested had the V617F JAK2 mutation and 11/12 over-expressed PRV1. Evaluation was performed at time of MPD diagnosis (n=9) or during evolution (n=36, median from diagnosis 43 months, range 7–135). Doppler transthoracic echocardiography (TTE) was used to estimate pulmonary artery systolic pressure (PASP), PAH being defined as PASP > 35 mmHg. Pulmonary function tests included slow vital capacity, total lung capacity, flow-volume curves, and lung diffusing capacity for carbon monoxide (DLCO and DLCOc when corrected for hemoglobin value). Gas exchange tests were performed using a treadmill exercise protocol allowing steady state alveolar-arterial differences measurement for rest and moderate exercise intensity.

Results: PASP could be estimated by TTE in 36 patients and was normal in all cases (mean 28 ± 3.5 mmHg). Pulmonary volumes and flow were normal in all MPD patients. However, we observed significantly decreased DLCOc in 17/28 patients (9/14 ET, 8/14 PV). Mean DLCOc was 89% ± 15 of predicted value (based on sex, age, height and weight) for the whole cohort (p=0.0004 by paired t-test), 84% ± 16 in ET (p=0.005) and 93% ± 12 in PV (p=0.035). Dead space to tidal volume ratio (VD/VT) at exercise was significantly increased in 19/28 patients (10/14 ET, 9/14 PV). Mean VD/VT was 28% ± 7, 28% ± 9 and 27% ± 5 in the whole cohort, ET and PV patients respectively, and was 18% ± 4 in controls (p=0.001 for the 3 groups). A negative correlation was found between VD/VT and DLCO (r=0.47, p=0.02). In PV patients, no correlation was found between hemoglobin value and DLCOc (p=0.37) or VD/VT (p=0.87).

Conclusion: Contrary to previous reports, in this prospective study in unselected MPD patients, no case of PAH was found. Yet, pulmonary function studies revealed in two third of MPD patients anomalies suggesting presence of pulmonary vascular disease a minima, with increased dead space at exercise correlated with decreased DLCO. These anomalies were similarly observed in PV and ET, and could be detected at diagnosis or during evolution in patients with normalized cell counts. A possible mechanism could be occlusion of the pulmonary micro-vascular bed, a phenomenon well described in terminal arterial circulation of MPD patients. Longer follow-up will show if abnormal dead space and DLCO are predictive of overt PAH development in MPD patients and could justify preventive measures.