Pulmonary hypertension during exercise underlies unexplained exertional dyspnoea in patients with Type 2 diabetes

Abstract Aims To compare the cardiac function and pulmonary vascular function during exercise between dyspnoeic and non-dyspnoeic patients with Type 2 diabetes mellitus (T2DM). Methods and results Forty-seven T2DM patients with unexplained dyspnoea and 50 asymptomatic T2DM patients underwent exercise echocardiography combined with ergospirometry. Left ventricular (LV) function [stroke volume, cardiac output (CO), LV ejection fraction, systolic annular velocity (s′)], estimated LV filling pressures (E/e′), mean pulmonary arterial pressures (mPAPs) and mPAP/COslope were assessed at rest, low- and high-intensity exercise with colloid contrast. Groups had similar patient characteristics, glycemic control, stroke volume, CO, LV ejection fraction, and E/e′ (P > 0.05). The dyspnoeic group had significantly lower systolic LV reserve at peak exercise (s′) (P = 0.021) with a significant interaction effect (P < 0.001). The dyspnoeic group also had significantly higher mPAP and mPAP/CO at rest and exercise (P < 0.001) with significant interaction for mPAP (P < 0.009) and insignificant for mPAP/CO (P = 0.385). There was no significant difference in mPAP/COslope between groups (P = 0.706). However, about 61% of dyspnoeic vs. 30% of non-dyspnoeic group had mPAP/COslope > 3 (P = 0.009). The mPAP/COslope negatively predicted V̇O2peak in dyspneic group (β = −1.86, 95% CI: −2.75, −0.98; multivariate model R2:0.54). Conclusion Pulmonary hypertension and less LV systolic reserve detected by exercise echocardiography with colloid contrast underlie unexplained exertional dyspnoea and reduced exercise capacity in T2DM.