Purpose: Recently, it has become increasingly recognized that pulmonary hypertension (PH) is a particularly threatening result of left-sided heart disease. However, there are few investigations about the impact of CPX variables on PH in dilated cardiomyopathy (DCM). The purpose of this study was to evaluate the ability of crucial cardiopulmonary exercise testing (CPX) variables to detect elevated pulmonary arterial pressure (PAP) and prognostic value in patients with DCM.
Methods: Ninety subjects with DCM (52±13 years, 69% male) underwent cardiac catheterization and CPX in our hospital. Predicted peak VO2 and the percentages of the predicted values achieved were calculated. Receiver operator characteristic (ROC) analysis was performed to assess the utility of CPX variables to distinguish between the presence and absence of PH. We followed all patients for the incidence of cardiac events for a mean of 4.3 years. Cardiac events were defined as cardiac death and hospitalization due to worsening heart failure.
Results: Overall mean values were: mean PAP (mPAP), 18.0±9.6 mmHg; plasma brain natriuretic peptide, 233±295 pg/mL; and left ventricular ejection fraction, 30.2±11.0%. Patients were divided into 2 groups on the basis of mPAP, namely DCM without PH group (mPAP<25 mmHg; n=75) and DCM with PH group (mPAP ≥25 mmHg; n=15). Peak VO2 and VE/VCO2 slope were significantly lower and higher in DCM with PH than in DCM without PH (11.3±3.6 mL/kg/min, 19.2±4.8 mL/kg/min, P<0.0001, 38.6±9.7, 29.2±7.4, P=0.002, respectively). The incidence of cardiac events were significantly higher in DCM with PH than in DCM without PH by the Kaplan-Meier method (P=0.002). After adjustment including the presence of PH, multivariate Cox proportional hazard analysis revealed that peak VO2 was the only significant independent predictor of cardiac events (odds ratio, 0.876; 95% confidence interval, 0.808 to 0.950). A cut-off value of percentages of predicted peak VO2 of 52.5% was the best predictor of a mPAP ≥25 mmHg by the ROC analysis (area under the curve [AUC: 0.911]; 95%CI: 0.846-0.977, P<0.001). VE/VCO2 slope >31.01 added significant diagnostic value (AUC: 0.800; 95%CI: 0.686-0.913, P=0.001).
Conclusions: Lower peak VO2 and higher VE/VCO2 slope were strongly associated with the presence of PH in patients with DCM. Moreover, peak VO2 was an independent predictor of cardiac events in patients with DCM. Taken together, CPX variables could have an important diagnostic utility for PH and provide prognostic information in patients with DCM.