Preliminary findings from a risk stratification, quality of life and burden of illness in pulmonary arterial hypertension pilot study

Abstract

Abstract Funding Acknowledgements Type of funding sources: Private grant(s) and/or Sponsorship. Main funding source(s): Janssen Australia and New Zealand (Janssen-Cilag Pty Ltd) Pulmonary Arterial Hypertension (PAH) management strategies focus on lowering mortality risk, extending the patient’s life span through the control and alleviation of disease symptoms, improving Quality of Life (QoL) and decreasing the burden of illness. This paper describes preliminary results of an Investigator Initiated pilot study exploring the relationship between PRO measures (PAH-SYMPACT, EQ5D5L); Burden of Illness assessment; and the ESC/ERS Risk Stratification. Findings Measures of PAH risk and QoL were assessed in 42 patients presenting to the IPAH clinic for management; mean age 69.2 [range 39, 84]; females 35 (83.3%), males 7 (16.7%). Tools analysed and compared included: PAH-SYMPACT and EQ-5D-5L as PRO and QoL measures; and ESC-ERS PAH risk stratification criteria to assess mortality risk. Health utility scores were derived from existing value sets: there was a 20% median reduction in QALYs for our patient population, with detrimental effects being more marked in those patients with greater impairment of EQ-5D-5L physical and cognitive-emotional domains (respectively: r = -0.92, p<0.001; and r = -0.85, p<0.001), PAH-SYMPACT physical domain reduction (r = -0.59, p<0.01), and PAH-SYMPACT cardiovascular symptoms (r = 0.64), p<0.01) and cardiorespiratory (r = 0.64, p<0.01) involvement. Physical domain scores were highly correlated for the two PROs studied (r = 0.74, p<0.001), but an unexpected poor correlation was seen in the cognitive/emotional domain (non-significant negative correlation) which requires further exploration. Unsurprisingly, EQ-5D-5L Visual Analogue Score for well-being was inversely correlated with physical domain impairment (r = -0.59, p<0.01), although it was noted that this did not apply for the cognitive/emotional domain. The agreement between tools was variable, with highest concordance seen for assessment of the physical domain (r = 0.74, p<0.001), and this raises the need to further define the merits of disease-specific (PAH-SYMPACT) vs generic (EQ-5D-5L) PRO analytic approaches. Tools examined in this study were well-correlated with symptomatic impairment, but neither PRO tool correlated with established risk markers as characterized in compound prognostic guides such as ESC-ERS criteria. Conclusions The fact that PRO indices showed little correlation with established risk stratification markers implies that the information provided by PROs is not redundant, and it is yet be assessed whether the incorporation of PRO’s may add further to the precision of risk assessment, a little-studied area which this group is further exploring. There should be more widespread and uniform use of PRO measures as part of standard PAH management.