Differentiating between circulatory versus non-circulatory causes of exercise intolerance in HFpEF using in vivo 31P magnetic resonance spectroscopy of exercising leg muscle

Heart failure with preserved ejection fraction (HFpEF) is characterized by severe limitations in exercise capacity. Identifying the etiology of exercise intolerance in HFpEF remains challenging due to heterogeneity of the disease and the presence of several comorbidities. This trial aimed to determine whether we could differentiate between circulatory versus non-circulatory causes of exercise intolerance using 31 phosphorus magnetic resonance spectroscopy (31P MRS) exercise testing. This observational trial included 18 HFpEF patients and 6 healthy controls who performed a supine, in-magnet cardiopulmonary exercise test using a prototype LODE MR-compatible cycle ergometer with an incremental workload protocol. Non-invasive in vivo 31P MRS measurements were collected from the quadriceps muscle to quantify dynamics in oxidative skeletal muscle metabolism. Primary outcome measures were end-exercise inorganic-phosphate/phosphocreatine ratio (Pi/PCr) and intramuscular pH. Evidence of circulatory-limited exercise intolerance was defined as increased end-exercise Pi/PCr ratio or decreased intramuscular pH (cut-off values 2.0 and 6.8, respectively). Patients with HFpEF were 50% female, aged 73 [64 - 76] years, had a mean BMI of 28.0 [25.0 – 29.6] kg/m2. Median LVEF was 51% [48 – 54%] and median NT-pro BNP levels were 626 [343-1931] pg/mL. Control patients were also 50% female, had a median age of 58 [56 - 65] years and a BMI of 22.7 [21.9 – 25.7] kg/m2. Nine patients in the HFpEF cohort exhibited circulatory insufficiency to support skeletal muscle oxidative metabolism, as indicated by acidification of the quadriceps muscle during exercise (Figure). In a subset of 5 HFpEF patients, no evidence for circulatory failure was found despite achieving maximum workload and exertion scores. Four patients were not able to reach a sufficient level of exercise intensity, which could correspond to a pattern of deconditioning. In controls, quadriceps Pi/PCr ratio increased to 1.1±0.8 while intramuscular pH remained neutral during exercise. This pilot study suggests that dynamics in skeletal muscle metabolism measured by 31P MRS can be used to differentiate between circulatory versus non-circulatory causes of exercise intolerance in HFpEF. Future research is needed to evaluate whether these patterns in oxidative skeletal muscle metabolism could help to unravel the heterogeneity in HFpEF. End-exercise parameters